This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Speech CODEC is the one that responsible on giving the analog original input output to the receiver digitization. But the process need to fulfill the nyquist frequency which the input signals must be band limited with its highest frequency as half of the inverse sampling period.
Analog to Digital Converter plays the role on digitizing the signal after it have been prepared in the sample and hold. The voltage level will be translated into a binary representation, and the process is also known as Pulse Code Modulation. Quality of the analog to digital conversion depends on the voltage levels. Its calculated by 2 to the power of number of bits (quantization bits).
Speech CODEC consists of Sample and hold circuit, Analog to Digital Conversion (ADC), Digital to Analog Conversion (DAC), and Filtering (Sample and Hold Reverse Operation).
In sample and hold after the original have been received, at this stage, it will takes a time variant voltage signal and divided into parts where only the constant voltage is remains in a period of time, this is call as a sampling period. This operation prepares the signal for
Figure 2: The complete conceptual A/D converter 
The signal has now been digitized and now it consist of sequence of numbers, sampling frequency, and also the number of quantization bits. Digital synthesizers are the one who will generate the number sequence that represent every voltage level, assign the quantization number and the sampling frequency.
Figure 3: Conversion from continuous time to discrete time results in quantization of the input signal. 
In order to hear the sound from a digital source it needs to first be transformed into an analog signal. This task is performed by the Digital to Analog Converter (DAC) taking a digital sequence of numbers as input and delivering a stepwise voltage signal. The number of bits and the speed of conversion limit the DAC.
After it has been quantized and now in the form of digital source, it need to turn into an analog signal for sound to be heard. This can be done by using the Digital to Analog Converter (DAC). The conversion works by referring to the sequence of numbers and deliver it into a stepwise voltage signal. The number of bits and speed of conversion limits the DAC.
The signal must now be recover and be smoothen after it have coming out from the DAC. So a Low band pass filter is requires to eliminates the components that are over the nyquist frequency o perform the smoothing. The final result is an analog signal.
DSP in Medicine
Hearing aid works like this; microphone will pick the sound and will transform it into an electric signal and will then be digitized. The digitized signal is then being filtered and selectively amplified to the frequency a band that suits the level of the patient hearing loss. The process also include in lowering the amplitude of the noise so that it will avoid excessive loudness to the ear. This works by having a certain threshold values and the received amplitude sounds exceed it. Then, it will be converted back to analog signal by (digital to analog converter) and delivered to the ear through the earphone.
Figure 4. In a further stage of signal processing in hearing aids, the amplitude-range may be compressed. Thus the gain of the system is progressively reduced when the sound-volume exceeds a specified level, as is the case near the end of this recording. A small delay in adapting the gain to the signal-levels may also be noted, as the average of the most recent amplitudes drives the gain control. 
The Acquisition of a blood pressure signal:
A blood pressure signal medically consists of components that made up about 20Hz of frequency. But when it is measured, the signal is always got contaminated by a noise signal that the main frequency is about 50Hz and other noise as well also present. So a sampling rate is required. To determine to sampling rate required we have to consider the maximum frequency presents instead of maximum frequency of interest. Since the highest frequency present is 50Hz so the sampling frequency has got to be above 10Hz. Aliasing will occur if the sampling rate is below 100Hz because the main (and other) noise will get mixed up with preferable frequency band of a blood pressure signal. In order to reduce the sampling rate, the noise have to be removed prior to sampling, using a low-pass (anti-aliasing) filter, and the sampling rate could be achieve to 40Hz with a cut-off frequency of 20Hz. This was carried out below and a sampling rate of 67Hz was chosen.
Figure 5. A recording of arterial blood pressure, contaminated by mains noise (50Hz) .
IIR filters to extract alpha activity from EEG signal
Electroencephalography (EEG) is the recording of electrical activity along the scalp. EEG measures voltage fluctuations resulting from ionic current flows within the neurons of the brain. In clinical contexts, EEG refers to the recording of the brain's spontaneous electrical activity over a short period of time
EEG is widely used in the diagnosis and investigation of epilepsy. Commonly a normal adult would have a noticeable which is alpha-rhythm; it is an oscillation of signal that strongly can be feel at back of the head that ranges from 8-13Hz, and they are at rest with the eyes closed. IIR filter are used in order to double the alpha-rhythm activity in EEG signal that contains about 50HZ interference, artifact and noise.
ADAPTIVE ARRAY IN SMART ANTENNA TECHNOLOGY
Smart antenna is the most efficient technology which is so helpful in providing a higher capacity in wireless network by reducing multipath and co-channel interference effectively. The smart antenna is able to adjust itself in order to changing the traffic conditions or signal environment and focus to the radiation for the desired direction. The set of radiating elements that employed by the smart antenna is in the form of an array which allow the beam pattern to move and switch and follows the desired user. The digital beamforming is the process where the signal and the radiation focused in a peculiar direction is combined.
First time, smart antenna system was designed to use in military applications to avoid the interfering or jammed signals form the enemy. The suppression of the interference was then implemented to personal wireless communication where interference is used to control the number of users in a network where they can be accommodated.
There are two approaches to the smart antenna to change their antenna pattern dynamically to mitigate the interference and multipath affects while coverage and the range increased. They are switched beam and adaptive array. The smartest approach between these two is the adaptive array system. It allows to tracks the mobile user continuously by control the main beam towards the user and at the same time forming nulls in the direction of the interfering signal as shown in figure below. A weight multiply the receive signal from every spatially distributed antenna elements. The weights are intricate in nature and regulate the amplitude and phase. An array output is yield when these signals are combined. These intricate weights are computed by a complicated adaptive algorithm, which is preprogrammed into the digital signal processing unit that administer the signal radiated by the base station.
Figure 9: Beam formation for adaptive array antenna system
By implementing advanced signal processing techniques offer a greater performances improvement where the information obtained is process by the antenna arrays. As comparison with switched beam systems, the adaptive array system is smarter as it is able to react to the changing RF environment dynamically. With the multitude of radiation patterns that they have compared to switch beam system where the fixed finite patterns which adapt to the everchanging RF environment. The antenna rays use in adaptive array just like switched beam system, is controlled by the signal processing. This signal processing controls the radiation beam toward the desired mobile user. This means it follows the users as they move, and understate interference arising from other users by introducing nulls in their directions at the same time. The figure below illustrated the beam formation for adaptive array antenna system.
Figure 10: Illustrated Beam Formation for Adaptive Array Antenna
The smart antenna system performs with the following functions:
The directions of arrival of all the incoming signals are estimated including the interfering signals and the multipath using the Direction of Arrival algorithms.
Desired user signal is recognized and detached from the rest of the unwanted incoming signals.
A beam is controlled in the direction of the desired signal and the user is traced as they moves while placing nulls at interfering signal directions by constantly updating the complex weighs.
The main beam radiation's direction in an array relies upon the phase difference between the elements of the array. Thus, it is feasible to continuously control the main beam in any direction by regulate the progressive phase difference Î² between the elements. The basis in adaptive array systems is forms using the same concept which the phase is regulate to attain maximum radiation in the desired direction.
Figure 11: Smart Antenna Architecture
A single desired beamformed output is formed when the signals incident at the individual elements are combined in a beamforming network. Then, they are brought to baseband or intermediate frequencies (IF's) before the incoming signals are weighted. The frequency down conversion is made at the receiver which has been provided at the output of each element. Digital Signal Processor which used in the adaptive antenna array systems weighted the incoming signal. The down-converted signal is required to be converted into digital format before being processed by the DSP which going to be done by the analog-to-digital converter (ADC). An accurate translation of the RF signal is required to be provided from the analog to the digital domain for accurate and better performances. Digital Signal Processor accepts the IF signal in a digital format and the software will process the digital data when driven to it. The processor translates the incoming data information and the complex weights is determined and multiplied to each element output to optimize the array pattern. In order to optimized, the contribution from noise and interference need to be minimize while the maximum beam gain is produced at the desired direction.
There are few benefits of Smart Antenna Technology:
Co-channel interference is reduced
Range of the base stations can be improved
The capacity of number of users increased.
Transmitted power is reduced.
The cells at the base station do not need to split the cells since the capacity increased therefore handoffs is reduced.
Mitigation of multipath effects
Compatible with any modulation method and bandwidth or frequency band.
DSP IN MOTOR CONTROL
At first, the motor control was invented by using the analog components as it is easy to design and able to be implemented by using inexpensive components. Nevertheless, there are several disadvantages in using the analog system such as the aging and temperature which brought to the condition of the components change and thus need regular adjustment. As many components get broken and adjustment process increases, the reliability of the system decreases. The endurance issues increases and the modification for upgrading the analog system become difficult as the design is hardwired.
In order to improve the design, the digital system is introduced instead of using the analog designs. By implementing the digital system design, the drifting is eliminated after almost all functions are executed digitally, while upgrading can be made easily using software and broken components is also decreasing as several functions is handle on chip by digital system.
The system costs are also decrease due to the high resolution; high speed and sensorless algorithm are provided by the Digital Signal Processor. Calculation processes become more important as precise control for better expenditure or radiation performances are provided. The DSP speeds-up calculations conclude the use of one cycle multiplication and addition instructions.
The main purposes of using the DSPs for motor control are:
The fixed point DSPs cost is lesser compared to the floating point DSP.
A 16 bits dynamic range is enough for almost applications. In case the dynamic range needs to be increased in a fixed point processor, it can be done by floating point calculation by the software.
The benefits if the DSP controllers are as follow:
Reducing the system cost with an efficient control in all speed ranges with implementing right dimensioning of power device circuits.
Torque ripple is reduced by performing high level algorithm which produces a lower vibration and long life span.
Reduction of harmonics is enabled by enhancing algorithms, in order to make easier requirements and reducing the filter cost.
Sensors' speed and position is removed after the sensorless algorithms are implemented.
The amount of memory required is reduced when the number of look-up tables is decreased.
A better performance is achieved after real-time generated of smooth near-optimal reference profiles and the trajectories moved.
A high-resolution PWM outputs is generated and power switching inverters is controlled.
A single chip control system is provided.
Figure 12: DSP Controller Architecture
DSP IN RADIO DETECTION & RANGING (RADAR)
Nowadays, DSP has been implemented in a lot of technologies especially in the production of Radar. These technologies are as follows:
Synthetic Aperture Radar
Ground Penetrating Radar
Air Traffic Control Radar
Weather Prediction Radar
The block diagram of a Modern Radar can be seen as below:
Figure 13 : Modern Radar Block Diagram
DSP helps a lot in Radar. It helps to combine the information, forming tracks, resolving ambiguities in range or Doppler measurements, Clutter Mapping grouping, managing the time and power and countering interference.
The signal detection in Radar is made by using the Clutter rejection technique. These techniques include are involving the two line delay cancellor, three line delay cancellor and a transversal filter.
Figure 14: Signal Detection Technique
The Moving Target Detector (MTD) signal processor is used in the Radar signal processing. The functions of the MTD are:
Detect the phase of input signal from IF circuits
Select the radar echoes based on Doppler properties
Analyses the properties of scanned radar horizon from the antenna resolutions
Suppressing the undesirable echoes from meteo clutter, intentional synchronous and asynchronous jamming and angel echoes.
Detect target on the basis of evaluating target contrast on background.
The block diagram of the Moving Target Detection (MTD) signal processor is illustrated below:
Figure 15: Moving Target Detection Block Diagram
The detection of signals in radar also is based on the adaptive thresholding and automatic detection. From the observations, the reference signal is internally generated which generate more sensitive and faster thresholds.
Digital beamforming is also important in radar. Phased array consists of small antenna elements with a phase shifter behind each element. Behind each antenna elements in digital arrays, there is an A/D converter and the beam steering. This A/D converter and the beam steering are performed by digital signal processors. The process of beam steering is executed by multiplying the signal from each array element by a complex weight.
There is variety of DSP technology has been grown over decade. The application of DSP has become major requirement in many areas such as communication, industrial products, audio video devices and medical. These technologies have variety approaches to employ DSP, from the used of Digital Signal Processor and General Purpose Processor. In comparison to find which hardware approaches is the best, a few comparison in term of speed, cost and energy efficiency should be made before develop the system.
Digital Signal Processor(DSPs)
General Purpose Processor
Specialized hardware performs all key arithmetic operation in 1 cycle
Multiplies and Shift take more than 1 cycle
Hardware support for managing numeric fidelity; shifters, guard bits, saturation
Others operation; saturation, rounding typically take multiple cycles.
Specific and complex instructions
Multiple operations per instruction
mac x0,y0,a x:(r0)+
Usually one operation per instruction
Von Neumann architecture
2-4 memory access/cycle
Typically 1 access/cycle
No caches-on-chip SRAM
Might use caches
Dedicated address generation units
Often, no separate address generation unit
Specific addressing mode ; auto increment, Modulo (circular), Bit-reversed (for FFT)
General-purpose addressing modes
Immediate data support
In term of speed comparison, parallelism in in how many parallel operations can be executed per cycle for both processor hardware technologies is taken into account. By increasing the clock speed also improve the performance of the processor to execute the operation. DSP processor common signal processing algorithms is the FIR and IIR filter. Both filter algorithm comparison is stated in the figure below.
More slower to implement using fixed-point arithmetic
Difficult to control
Derived from analog
No analog history
No limited cycle
Digital Signal Processor(DSPs)
General Purpose Processor
Less power because it is Fixed-point processors
Higher power consumption for high-performance General Purpose Processor
Energy efficiency is determined on hardware implementation; it based fabrication process, voltage management, circuit design and logic design. Back to hardware architecture, memory usage also influences the usage of energy in the processor. It depending on the matches between the instruction set and the task that needs to be handling. Other than code or instruction set quality, compiler quality also influence the energy efficiency of these DSP technology.
Digital Signal Processor(DSPs)
General Purpose Processor
Due to less hardware complexity cost for producing is less
Complexity and more expensive for development and producing.
To achieved low-cost and higher performance input and output processor. The development of DSP processor need to concentrate the die size and it include the fabrication process, size of on-chip memory and on-chip peripherals.
The Evolution of DSP Processor, By Jennifer Eyre and Jeff Bier, Berkeley Design Technology, Inc. (BDTI)
FFT BENCHMARKING FOR DIGITAL SIGNAL PROCESSING
TECHNOLOGIES, F. Stefani1, A. Moschitta1, D. Macii1, D. Petri2 Department of Information and Electronic Engineering, University of Perugia, Perugia, Italy