The Technique Of Power Control Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Power control is a technique for reducing energy consumption and can also reduce interference and improve spatial re use of wireless channels. [Shu-min chen et al]. Power control has been proposed based on an RTS-CTS handshake in the context of IEEE 802.11 [S. Agarwal et al, J. Gomez et al ]. Asymmetric links is being introduced with different power levels among different nodes. Therefore in the proposed protocol for power control scheme, RTS and CTS are transmitted using the highest power level and DATA, ACK are transmitted using the minimum power level that is necessary for the nodes to communicate. This chapter illustrates that the technique proposed scheme above has a shortcoming that increases collision and degrades network throughput {Eun-sun Jung, 2006]. A new power control protocol that does not grade throughput is being presented.

M.B Pursley et al proposed a power control mechanism that can be incorporated into the IEEE 802.11. RTS-CTS handshake which allow a node A to specify its current transmission power level in the transmitted RTS and allow receiver node B to include a desired transmission power level in the CTS sent back to A. Receiving the CTS, node A then transmits DATA using the power level specified in the CTS. This mechanism allows B to help A choose the appropriate power level, so as to maintain a desired signal-to-noise ratio.

PARO, a power- aware routing optimization proposed by J.Gomez et al consumes low energy and also chooses a cost function based on the transmission power level at each hop on a route to determine a low energy- consuming route between a pair of nodes. PARO uses a similar protocol known as the BASIC protocol that can be compared to the power control MAC Protocol. Basic protocol shows that stations can use the maximum power to transmit RTS and CTS frames, and the minimum required power to transmit DATA and ACK frames. The RTS-CTS handshake is used to decide the minimum required power for the subsequent DATA-ACK transmissions.

S. Agarwal et al presented a power control protocol that is similar to the BASIC scheme, which maintains a table for the minimum transmission power necessary to communicate with neighbour nodes. It also allows each node to increase or decrease its power level dynamically. Collisions occur when different power levels among nodes result in asymmetric links.

S - L. Wu et al propose power control protocol that uses one control channel and multiple data channels. A control channel is used to assign data channels to nodes. An RTS, CTS, RES a special packet), and broadcast packets are transmitted through the control channel using the highest transmission power. Using the RTS-CTS handshake, source and destination nodes decides which channel and what power level to use for data transmissions, At the signal of CTS, the source sends and RES to the destination to reserve a data channel. Then, DATA and ACK transmission occur on the reserved data channel using the negotiated power level from the CTS-CTS handshake.

J-P. Ebert et al also propose scheme that is base on the observation that reducing transmission power can result in energy savings but can also result in more errors. Higher bit error rate can lead to increased re transmission, consuming more energy. The protocol by J.P Ebert et al chooses an appropriate transmission power level based on the packet size.

IEEE 802.11 results in performance degradation for nodes that uses lower transmission power than their neighbour nodes. Poojary et al propose a scheme to improve the fairness

Power controlled Multiple Access (PCMA) protocol was proposed by J.P. Monks et al allows different nodes to have different transmission power levels. PCMA uses two channels, one for "BUSY TONE" and the other for all other packets. Power controlled multiple Access uses busy tone in place of RTS-CTS to overcome the hidden terminal problem. And node receiving DATA packet sends a busy tone periodically. Power level at which bust tone is being transmitted by a node is equal to the maximum additional noise the node can tolerate. Any node wishing to transmit a packet first waits for a fixed duration and senses the channel for busy tones from other nodes. Strength signal of busy tones received by a node is utilized to determine the highest power level at which this node may transmit without interfering with other on-going transmissions.

Power control is basically used for the purpose of topology control and used to establish energy efficient spanning tree for multicasting and broadcasting [J.E. Wieselthier et al].


Two medium access protocols are being specified for the IEEE 802.11 which are PCF (Point Coordinate Function) and DCF (Distributed Coordinate Function). PCF is a centralized scheme and DCF is a fully distributed scheme. In so doing, DCF is being considered in this dissertation.

Three terms need to be address in the MAC protocol that will be used in the rest of this chapter.

Transmission range

Carrier sensing range

Carrier sensing zone

Transmission Range: When a station A is within the transmission range of another station B, Station A can receive and correctly decode frames from station B.

Carrier Sensing Range: Station in the carrier sensing range can sense the sender's transmission. Usually, the carrier sensing range is larger than the transmission range ( a typical assumption is that the radius of the former is twice larger than that of the latter(Eun Sun Jung, 2006)). The transmission range and the carrier sensing range also depend on the sender's transmit power level.

Carrier Sensing Zone: its defined as they are of the carrier sensing range excluding the transmission range. A station within the carrier sensing zone of the transmitted can only sense the signal but cannot decode the transmitted data correctly. The diagram below illustrate these definitions

Diagram 16

The diagram above shows the transmission range, carrier sensing range and carrier sensing zone for node C. When node C transmits a packet, B and D can receive and decode it correctly since they are in the transmission range. However, A and E can only sense the signal and cannot decode it correctly because they are in the carrier sensing zone.

Distribution coordinate function in IEEE 802.11 is based on CSMA/CA (Carrier sense multiple Access with collision Avoidance). Carrier sensing is being done using physical carrier sensing (by Air interface) as well as virtual carrier sensing. Virtual carrier sensing uses the duration of the packet transmission, which is included in the header of RTS, CTS and DATA frame. The duration included in each of these frame can be used to infer the time when the source node would receive an ACK frame from the destination node [Eun-sun Jung, July 2006].

In the IEEE 802.11 MAC protocol, each station maintains a network allocation vector (NAV), which indicates the remaining time of the ongoing transmission sessions. When a station hears an RTS or a CTS frame, it will update its NAV using the duration information specified in the frame. Thus, when source and destination station transmit RTS,CTS and DATA frame, station in the corresponding transmission range can correctly receive these packet and update their NAVs. The channel is considered to be busy if their physical or virtual carrier sensing indicated that the channel is busy.

Diagram 17

The diagram above shows how nodes in transmission range and the carrier sensing zone adjust their NAVs during RTS-CTS-DATA-ACK transmission. SIFS, DIFS and EIFS are inter frame spaces (IFS) which are specified in IEEE 802.11.

IFS is the time interval between frames which defines four (4) IFS for IEEE 802.11. SIFS (Short inter frame space), PIFS (PCF inter frame space), DIFS (DCF inter frame space) and EIFS (Extended inter frame space). IFS provides priority levels for accessing the channel, SIFS is the shortest of the inter frame space and is used after RTS, CTS and DATA frame to give the highest priority to CTS, DATA and ACK respectively. Whenever a channel is idle in the DCF, a node waits for the DIFS duration before transmitting any packet.

When destination and source nodes transmit RTS and CTS, node in transmission range correctly receive these packets and set their NAVs for the duration of the whole packet transmission. Nodes within the carrier sensing zone only sense signal and can't decode correctly, so these nodes set their NAVs for EIFS duration. EIFS is used to protect an ACK frame at the source node. The main purpose of EIFS is to provide enough time for a source node to receive the ACK frame, so the duration of EIFS is larger than that of an ACK transmission [Eun-sun Jung, 2006].

IEEE 802.11, EIFS is obtained by using the SIFS, the DIFS and the length of time to transmit an ACK frame at the physical layer's lowest mandatory rate, given by the equation EIFS= SIFS+DIFS+ [(8*ACK size) + Preamble length + PLCPHeader length]/Bit Rate, where ACK size= length of an ACK frame (in byte) and Bit Rate is the physical layer's lowest mandatory rate. Preamble length is 144 bits and PLCPHeaderLenght is 48 bits [IEEE, wireless LAN Medium Access Control (MAC) and (PHY) specification, Nov 1997]. The IEEE 802.11 wireless LAN Mac and PHY specification states that the EIFS is used whenever the physical layer has indicated to the MAC that a frame transmission was begun but that frame transmission did not result in the correct reception of a complete MAC frame with a correct FCS (frame check sequence) value. EIFS interval begins by following the indication by the physical layer that the channel is idle after sensing of the erroneous frame.

Simulation carried out by Eun-sun Jung uses a somewhat conservation variation on the above 802.11 specification. Whenever a node senses a transmission but can't receive the transmission correctly, EIFS is used. IEEE 802.11 do not prevent collision completely due to a hidden terminal-nodes in the receiver's carrier sensing zone, but not in the sender's carrier sensing zone or transmission range can cause a collision with the reception of DATA packet at the receiver. Diagram........ Illustrated this further

Diagram 18

Suppose station C transmit a packet to station D,C and D transmit an RTS and CTS respectively, nodes A and F will set their NAVs for EIFS duration. When C transmits DATA, A defines its transmission because it sense any signal during C's DATA transmission, so it consider the document the channel to be idle. (F is in D's carrier sensing Zone, but not in C's). When F starts a new transmission, it can cause a collision with the reception of DATA at D. As F is outside D's transmission range by symmetry, D may be outside F's transmission range. Since F is in D's carrier sensing zone, it means that F can present sufficient interference at node D to cause a collision with DATA being received by D.


Several works try to investigate the Basic power control scheme and its limitation. S. Agarwal et al , J. Gomez et al, P. Karn and M.B. Pursley tries to look at the scheme and possible limitation of basic power control protocol.

4.3.1 Basic protocol Description

Energy consumption reduction can be done by power control which means power control may introduce different transmission power levels at different hosts, creating an asymmetric situation where a station A can reach station B, but B cannot get through to A. Power transmitted used at different stations may also result in increase collision unless adequate precaution are taken. An example is being illustrated in Diagram below where station A and B uses lower power than station C and D. When A transmits a packet to B, C and D may not sense the transmission. So, when C and D transmit as well to each other using a higher power, their transmission will collide with the on-going transmission from station A to B.

Diagram 19

The solution to this problem ( as a modification to IEEE 802.11) that was suggested by S. Agarwal et al , J. Gomez et al , P. Karn et al and M.B Pursley is to transmit RTS and CTS at the highest possible power level but transmit DATA and ACK at the minimum power level necessary to communicate. This is being referred to as the Basic Scheme.

Diagram 20

The diagram above illustrates the Basic scheme where station A and B sends RTS and CTS respectively with the highest power level so that node C receives the CTS and defers its transmission. Energy can be conserve by the nodes by using lower power for DATA and ACK.

RTS-CTS handshake in BASIC scheme is used to decide the transmission power for subsequent DATA and ACK packets. Let Pmax be the maximum transit power. Suppose station V wants to send a data packet to station U. Station V should use a power level of Pmax to send its RTS frame. When station U receives this RTS frame, it also replies to CTS frame at power level Pmax. Station V receives the CTS frame, it calculates the minimum requires power level, Pdesire = Pmax/Pr *Prmin * C, based on the receieved power level Pr of the CTS frame, where Prmin is the minimum necessary receieved sihnal strength and C is a constant. Then station V used power level Pdesired to transmit its DATA frame. Similarly, station U calculates its power level Pdesired to transmit its ACK frame [Shu-min Chen et al ].

Two assumptions are being derived from this method. Firstly, signal attenuation between the destination and source nodes is assumed to be in the same directions and secondly, the receivers noise level is assumed to be below some predefined threshold. This method can result in unreliable communication when the assumptions are wrong. Modification to the CTS format does not require this alternative. This alternative is being used by Eun-sun Jung in the BASIC simulation.

4.3.2 Deficiency of the Basic Protocol

The Basic protocol may degrade network throughput and even cause higher energy consumption. This is possible because RTS and CTS are sent using Pmax , DATA and ACK packet are sent using the minimum necessary power to reach the destination. When the RTS or CTS receives nodes from the neighbour, the NAVs for the duration of the DATA-ACK transmission is been set.

Diagram 21

This diagram illustrates the problem of Basic protocol.

Suppose station D wants to transmit a packet to node E. When D and E transmit the RTS and CTS respectively, B and C receives the RTS, and F and G receive the CTS, so these nodes will defer their transmission for the duration of the D - E transmission. Station A is in the carrier sensing Zone of D (where D transmit at Pmax) si it will only sense the signal and cannot decode the packet correctly. Station A will set its NAV for EIFS duration when it senses the RTS transmission from D. In the same way, station H will set its NAV for EIFS duration following CTS transmission from E.

When the power control transmission is not used, carrier sensing zone for RTS-CTS and DATA-ACK is the same since all packet are sent using the same power level. In the BASIC scheme, when a source and distribution pair decided to reduce the transmission power for DATA - ACK, the transmission range for DATA - ACK is smaller than that of RTS-CTS; in the same way, the carrier sensing zone for DATA-ACK is also smaller than that of RTS-CTS.

When station D and F reduces their transmission power for DATA and ACK transmission range are reduced. Thus, only C and F cab correctly receive the DATA and ACK packets respectively. Furthermore, since station A and H cannot sense the transmitting at the power level Pmax, this transmission causes a collision with the ACK packet at D and DATA packet at E. This Results in throughput degradation and higher energy consumption (because of retransmission).

In MAC protocol, IEEE 802.11 doesn't prevent station in the carrier sensing zone (station H in diagram 22) from causing collisions with the DAT packet at the destination (station E in the diagram 22). In the Basic protocol, station A cant sense D's DATA transmission at the lower level, so a transmission from A can interfere with the reception of the ACK at D. This indicates that Basic is liable to collision, degrading throughput as shown by MAC protocol. Basic scheme has been considered for saving energy by past researchers but its deficiencies were not identified. For example, P. Karn considers 100 x 100m^2 area for its simulation making every node to correctly decode RTS or CTS and will know the duration of the remaining packet transmission. The negative impact of the Basic power control was not really talked about.


The proposed Power Control MAC (PCM) is similar to the Basic scheme in the sense that it uses power level Pmax for RTS-CTS and the minimum necessary transmission power for DATA-ACK transmissions. PCM procedure is described below.

Source and destination nodes transmit RTS and CTS using Pmax. Carrier sensing zone nodes set their NAVs for EIFS duration when they sense the signal and cant decode it correctly.

Source node can transmit DATA using a lower level, similar to the Basic scheme.

Avoiding potential collision with the ACK, source node transmits DATA at the power level Pmax, periodically, for just enough time for node in the carrier sensing zone to sense it.

Destination station transmits ACK using the minimum required power to reach the source node that is similar to Basic scheme.

Diagram 22

Example of power adjustment by N. H. Vaidya et al based on the scenario in fig ......, notifying stations in the carrier sensing zone, the transmit power for the DATA frame is increased periodically to the power level Pmax.

The major difference between PCM and Basic Scheme is that the power level for transmitting DATA frames is periodically increased to the power level Pmax from the power level Pdesired. That is, the power level to transmit DATA frames is alternated between Pmax and Pdesire with a period of one EIFS. With this modification, other stations that may cause collision will periodically observe the existence of carriers and postpone their transmission. Since the transmit power for DATA frames is increased every EIFS duration, proper NAVs can be set at other stations. In addition, the length of the duration to transmit at the power level Pmax should be long enough for physical carrier sensing [N.H Vaidya and E. S Jung, 2002].

According to the IEEE, wireless LAN MAC and PHY specification, Nov 1997, 15 µs should be adequate for carrier sensing and the time required to increase output power from 10% to 90% of maximum power (or power down from 90% to 10% of maximum power) should be less than 2µs. Thus, it is believed that 20 µs should be enough to power up(2 µs), sense the signal (15 µs) and power down (2 µs).

In a simulation that was carried out by E. S Jung, EIF duration was been set to 212 µs using a 2mbps bit rate. A node transmit DATA at Pmax every 190 µs for a 20 µs duration in PCM. Thus, the interval between the transmissions at Pmax is 201 µs, which is shorter than EIFS duration. A source node starts transmitting DATA at Pmax for 20 µs and reduces the transmission power to a power level adequate for a given transmission which is shown in the diagram above. The node also transmits DATA at Pmax for the last 20 µs of the transmission.

This modification done by E. S Jung shows how PCM overcomes the problem of Basic scheme and can achieve throughput comparable to 802.11, but uses less energy. It should be noted that PCM like 802.11 doesn't completely prevent collisions. Specifically, the goal is to match the performance of 802.11 while reducing energy consumption.

Several research scheme has been proposed to improve on the PCM scheme develop but none could still perform adequately. Even though they all achieve the aim of saving energy, improve spatial channel reuse and demonstrate significant throughput increase over IEEE 802.11. They still lack behind in some areas such as introducing additional hardware cost and implementation complexity by requiring two transceiver and two channels.

A.Muqattash and M. Kranz came up with a throughput-oriented protocol which was name POWMAC (power control mac protocol) that follows the single channel and single transceiver design principle of IEEE 802.11. POWMAC uses an access window to allow a series ofRTS/CTS exchanges to take place before multiple concurrent data packet transmission start. Collision avoidance information attached in CTS is used to bound the transmission power of potential interfering nodes, rather than silencing those nodes. Therefore, spatial channel reuse is improved. POWMAC yield great increase in network throughput over IEEE 802.11, and reduction in energy dissipation.

Lujun Jia et al also proposed a new class of power control scheme called PCS that increases network throughput by using a novel transmission power function P(t) to compute am appropriate transmission power, so that better spatial channel reuse is achieved. PCS doesn't advertise no collision avoidance information like the POWMAC. Instead, node chooses a transmission power level based on its traffic distance d, and an estimate of the interference level it experiences. PCS indentifies a single scheme that shows significant improvement in network throughput, energy efficiency and fairness over 802.11 and adheres to the same single-channel and single transceiver design principle. One major problem of PCS scheme is that it doesn't integrate well within a resource efficient multi hop routing protocol.


In order to improve the deficiency of PCM, that is collision, Y. Mahmood et al proposed an Improved Power Control MAC protocol for wireless ad hoc network. The IPCM protocol is similar to the PCM scheme proposed ealier by E. S. Jung and V. H Vaidya except that the source node transmits DATA with the optimum level. The power level is periodically increased for just enough time not to Pmax in PCM but to a suitable level (Pai ) sufficient to avoid collisions. IPCM is an improved version on the PCM protocol. Y. Mahmood et al states that the PCM transmit data with maximum periodic pulse power which means reserving maximum transmission area for the giving ongoing transmission even the distance between the transmitter and receiver is small. The reason behind using maximum periodic pulse power is to increase the sensing range for informing the neighbour nodes about ongoing transmission in order to reduce the interference abd increase energy conservation. But, doing this affects the total throughput of the network since some nodes in the maximum carrier sensing range can also transmit data successfully to its corresponding receiver without affecting the first ongoing transmission.

Eun Sun Jung and V. H. Vaidya illustrated that the number of the interference nodes reduces in the chain topology with 30 flows using PCM scheme compared to the BASIC scheme. Using the IPCM protocol, all packets transmitted usning the optimim powers and with the help of the interference analysis, it was found out that optimim carrier sensing range lesser than the maximum is sufficient to avoid the collision. Which means that other concurrent transmissions can take place at the same time.

In an illustration stated by Y. Mahmood et al, that in a chain topology of 31 nodes with 30 flows and the distance between adjacent node pairs is 40m, the carrier sensing range of 134m is enough to avoid interference compared to 55m as the case maybe in PCM. Rtop will be the transmission range of the RTS using the optimum power. Because RTS and CTS uses the sameoptimum power, Rtop will represent CTS as well. Suppose the periodic pulse power is also the same (i.e Optimum), the carrier sensing range is at least twice time the transmission range [A. Kamerman and L. Montebau, 1997]. If the receiver is at the edge of the transimission range of the transmitter, carrier sensing range (Rcs) will cover both transimission ranges of RTS and CTS as shown below

Diagram 23

This may be consider as the worst scenario because the periodic pulse power will be greater than the optimum used power. It now depends on the distance between the transmitter and the receiver for any optimum power lesser than the maximum power. This means that the ongoing transmission will be completely covered by the sensing range. Any node in the sensing range but not in the RTS or CTS range will notice the transmission and defer its transmission request. According to the IEEE 802.11 DCF, this node will maintain a NAV that indicated the remaining time of the ongoing transmission session. When the transmitter finished the data transfer, a node in the carrier sensing range goes to a back - off period to sense the medium again. If any more data are to be transmitted by the receiver ACK, the node in the back - off mode will notice the medium busy and maintain another NAV.

The proposed IPCM works in the following basic steps:

Transmitter sends an RTS with the optimum transmit power level including the level of that power.

Receiver decodes the RTS, finds the power level value, observes the SIR (Signal Interference Ratio) value and sends the DATA with optimum power, and periodically increases the power level of the DATA packets to Pai to avoid interferences. The Pai value is selected based on the ration of the channel capacity and the carrier sensing range.

The receiver sends ACK using the optimum power level.

An algorithm was also proposed at the transmitter node and receiver's node. At the transmitter's node, the algorithm is used to check node address, decode when to send RTS, receives CTS and observe it receive power also, extract SIRvalue from CTS packet and determine the capacity_area_ratio according to the following equation:

Simulation carried out in the proposed IPCM showed that comparing the performance of IPCM scheme with the IEEE 802.11, BASIC and PCM power control schemes under different network topologies, different data rates and different packets sizes achieves more total data delivered per joule. It means that IPCM scheme can achieve a higher reduction in the energy consumption. Also, the simulation results show that IPCM scheme highly improves the network throughput compared to all other schemes that have being proposed in the past because it was designed to avoid interference (collision), save energy and improve throughput.


Several researchers in the past have studied a MAC protocol that uses the maximum transmission power for RTS-CTS and the minimum necessary transmission power for DATA-ACK with the aim of saving energy. This is being referred to as Basic Scheme. It has been shown that Basic scheme increases collisions and retransmission that results in more energy consumption and throughput degradation.

In IEEE 802.11, carrier sensing range for RTS-CTS is the same as that of DATA-ACK since transmission power doesn't change. Likewise in Basic, carrier sensing range for RTS-CTS and DATA_ACK vary because the transmission power can be different for those packets. So, when using Basic, nodes in the carrier sensing zone for RTS-CTS can cause collision with On-going DATA-ACK transmission because these nodes may not sense DATA transmission, which consumes more energy. This causes the Basic scheme to often yield an aggregate throughput worse than IEEE 802.11 without power control [E. S Jung, Aug 2006].

With this in mind, PCM (Power control MAC protocol) was proposed that periodically increases the transmission power during DATA transmission. E.S Jung simulation results shows that PCM achieves energy saving without causing throughput degradation.

A major blow for PCM is that it requires a frequent increase and decrease in the transmission power which makes the implementation very difficult. Another method is to replace this higher power level for data by a busy tone at Pmax in a separate channel, with one channel being used for the busy tone and the other for RTS-CTS-DATA-ACK. Fading may adversely affect the PCM performance and a variation of PCM, a different time interval can also be used between the transmission at Pmax during a packet transmission. This variation indicated that there exist a trade off between the performance and energy savings.

Knowing full well that PCM provides energy saving, it does not yield improved spatial reuse as compared to IEEE 802.11.The deficiency of spatial reuse was improved on the PCM as compared to the IEEE 802.11 brought about POWMAC and PCS that uses an access window to take place before multiple concurrent data packet transmission starts. POWMAC advertise collision avoidance information and PCS doesn't which make both protocol different. The Improved Power Control for MAC (IPCM) uses the source node to transmit DATA with optimum level. The power level increases periodically to a suitable level (Pai) to avoid collision and help attain higher reduction energy consumption, improve network throughput and in the same way, save energy just like the PCM.