Digital signal processing selection and introduction
DSP, or Digital Signal Processing (DSP), is an emerging discipline that covers many disciplines and is widely used in many fields. Since the 1960s, with the rapid development of computer and information technology, digital signal processing technology has emerged and developed rapidly. Digital signal processing is a method of processing real-world signals by performing transformations or extracting information using mathematical techniques, and these signals are represented by a sequence of numbers. In the past two decades, digital signal processing has been widely used in communications and other fields. Semiconductor manufacturers such as Texas Instruments and FREESCALE have strong capabilities in this field.
Second, digital signal processing selectionMainly consider processing speed, power consumption, program memory and data memory capacity, on-chip resources, such as the number of timers, the number of I / O ports, the number of interrupts, the number of DMA channels. The main suppliers of DSP are TI, ADI, Motorola, Lucent and Zilog, among which TI has the largest market share.
TI is now pushing four series of DSPs
1) C5000 series (fixed point, low power consumption): C54X, C54XX, C55X Compared with other series, the main feature is low power consumption, so it is most suitable for personal and portable Internet and wireless communication applications, such as mobile phones, PDAs, GPS and other applications. The processing speed is between 80 MIPS and 400 MIPS. C54XX and C55XX generally only have McBSP synchronous serial port, HPI parallel interface, timer, DMA and other peripherals. It is worth noting that the C55XX provides an EMIF external memory expansion interface that can be used directly with SDRAM, while the C54XX cannot be used directly. There are only two digital IOs in both series.
2) C2000 series (fixed point, controller): C20X, F20X, F24X, F24XX, C28x This series of chips has a large number of peripheral resources, such as: A/D, timer, various serial ports (synchronous and asynchronous), WATCHDOG, CAN Bus/PWM generator, digital IO pin, etc. It is a DSP optimized for control applications. Among all DSPs in TI, only C2000 has FLASH, and only this series has an asynchronous serial port that can be connected to the UART of the PC.
3) C6000 series: C62XX, C67XX, C64X This series is famous for its high performance and is most suitable for broadband network and digital imaging applications. 32bit, where: C62XX and C64X are fixed-point series, and C67XX is a floating-point series. This series offers an EMIF extended memory interface. This series is only available in BGA packages and can only be used to make multi-layer PCBs. And the power consumption is large. The VC33 in the C3X, which is also a floating-point series, is still widely used, but it is still widely used, but its speed is low, up to 150MIPS.
4) OMAP series: OMAP processor integrates ARM command and control functions, and also provides DSP low-power real-time signal processing capability, which is most suitable for mobile Internet devices and multimedia appliances.
Other series of DSPs have had scenery, but now they are not the main products of TI. In addition to the C3X series, others are basically in the elimination stage, such as: C3X floating point series: C30, C31, C32C2X and C5X series: C20, C25, Each of the C50's DSPs has its main application area.
Digital signal processing system
Third, the selection parameters of the digital signal processing chipAccording to the application and design goals, the focus of selecting DSP chips is also different. The main parameters include the following aspects:
(1) Operation speed: First, we must determine the algorithm of digital signal processing. After the algorithm determines the calculation amount and completion time, it is generally determined. According to the calculation amount and its time requirement, the lower limit of the DSP chip operation speed can be estimated. When selecting a DSP chip, the measurement speed of each chip is mainly as follows:
MIPS (Millions of Instructions PerSecond), millions of instructions / second, the general DSP is 20 ~ 100MIPS, the TMS320B2XX using the very long instruction word is 2400MIPS. It must be pointed out that this is a measure of the speed of the fixed-point DSP chip. It should be noted that the indicator provided by the manufacturer generally refers to the peak indicator. Therefore, the system should be designed with a certain margin. MOPS (Mill
ionsofOperationsPerSecond), performing millions of operations per second. The problem with this indicator is what is an operation. Usually, operations include CPU operations, address calculation, DMA access data transfer, and I/O operations. Generally speaking, the higher the MOPS means the faster the product-accumulation operation. MOPS can provide a comprehensive description of the performance of the DSP chip.
MFLOPS (MillionFloatingPointOperationsPerSecond), millions of floating-point operations / second, which is an important indicator of floating-point DSP chips. For example, when the TMS320C31 is at 40MHz, the processing power is 40MFLOPS, and the TMS320C6701 can achieve 1GFLOPS in single-precision operation when the instruction cycle is 6ns. Floating point operations include floating point multiplication, addition, subtraction, storage, and so on. It should be noted that the indicator provided by the manufacturer generally refers to the peak indicator. Therefore, the system should be designed with a certain margin.
MBPS (MillionBitPerSecond), which is a measure of the data throughput of the bus and I/O ports, that is, the bandwidth of a bus or I/O. For example, for TMS320C6XXX, 200MHz clock, 32bit bus, the bus data throughput rate is 800Mbyte/s or 6400MBPS.
ACS (Multiply-Accumulates PerSecond), such as TMS320C6XXX multiplying acceleration up to 300MMACS~600MMACS.
The instruction cycle, which is the time required to execute an instruction, is usually in ns (nanoseconds). For example, the TMS320LC549-80 has an instruction cycle of 12.5 ns at a frequency of 80 MHz. MAC time, the time it takes to perform a multiply and add operation: Most DSP chips can perform a MAC operation in one instruction cycle.
FFT/FIR execution time, the operation time of an N-point FFT or N-point FIR program. Since the FFT operation/FIR operation is a typical algorithm for digital signal processing, this indicator can be used as a comprehensive indicator for measuring chip performance.
(2) Operation accuracy: Under normal circumstances, the operation precision of the floating-point DSP chip is higher than that of the fixed-point DSP chip, but the power consumption and price also increase. Generally, the fixed-point DSP chip has a word length of 16 bits, 24 bits or 32 bits, and the floating point chip has a word length of 32 bits. Accumulators are typically 32 or 40 bits. The fixed-point DSP is characterized by high frequency, high speed, low cost and low power consumption. It is mainly used in the fields of control, communication, voice/image, consumer electronics and other fields with low complexity. It is usually possible to solve problems with fixed-point devices, using fixed-point devices as much as possible because of its economy, speed, low cost, and low power consumption. However, when programming, pay attention to the dynamic range of the signal, and add a scaling operation that limits the dynamic range of the signal. Although we can improve the accuracy of the algorithm by improving the algorithm, this will increase the complexity and computation of the program. The speed of floating-point DSP is generally lower than that of fixed-point DSP. Its cost and power consumption are higher than fixed-point DSP. However, because it adopts floating-point data format, the processing precision and dynamic range are much higher than fixed-point DSP, which is suitable for operation. Applications with high complexity and high precision requirements; even for general applications, when programming floating-point DSPs, there is no need to consider data overflow and insufficient precision, so programming is more convenient and easier than fixed-point DSP. Therefore, the operation accuracy requirement is a trade-off problem, and it is necessary to determine an optimum combination point based on experience and the like.
(3) Choice of word length: Generally, floating-point DSP chips use 32-bit data words, and most fixed-point DSP chips are 16-bit data words. Motorola's fixed-point chips use 24-bit data words to achieve a trade-off between fixed-point and floating-point accuracy. The word length is an important factor affecting the cost. It affects the size of the chip, the number of pins, and the size of the memory. When designing, the minimum data word is selected as much as possible.
(4) On-chip hardware resource arrangement such as memory: including the size of the memory, the number of on-chip memories, the bus addressing space, and the like. The size of the on-chip memory determines the speed and cost of the chip. For example, TI's same series of DSP chips, different types of chip memory configuration and other hardware resources are different. Through careful analysis of the algorithm program and application goals, the requirements for the on-chip resources of the DSP chip can be roughly determined. Several important considerations are the number of on-chip RAM and ROM, the availability of external memory, the bus interface/interrupt/serial port, etc., whether or not there is A/D conversion.
(5) Development and debugging tools: Perfect and convenient development tools and related support software are necessary conditions for developing large-scale and complex DSP systems, which play an important role in shortening the product development cycle. Development tools include software and hardware. Software development tools mainly include: C compiler, assembler, linker, library, software simulator, etc. After determining the DSP algorithm, the written program code is simulated by the software simulator to determine the necessary performance indicators. Hardware development tools include online hardware simulators and system development boards. The online hardware emulator is usually a JTAG peripheral scanning interface board, which can debug the designed hardware in-line. Before the hardware system is completed, the DSP software of the designed function can be run in real time on different development boards to improve the development efficiency. Even in a small number of products, the development board is directly regarded as the final product.
(6) Power consumption and power management: In general, personal digital products, portable devices, and outdoor devices have special requirements for power consumption, so this is also a consideration. It usually includes power
Voltage selection and power management functions. The supply voltage is generally low, and the low-voltage power supply of the chip is usually 3.3V, 2.5V, 1.8V, 0.9V, etc. At the same clock frequency, their power consumption will be much lower than that of the 5V supply voltage. . After the management of the power supply is strengthened, the power consumption is usually saved by using a sleep mode or a standby mode. For example, TI provides detailed application instructions that vary with the type of instruction and processor configuration.
(7) Price and manufacturer's after-sales service factors: The price includes the price of the DSP chip and the price of the development tool. If an expensive DSP chip is used, even if the performance is high, its application range is definitely limited. However, low-cost chips must have fewer functions, less on-chip memory, and poorer performance, which brings certain difficulties to programming. Therefore, according to the actual system application, determine a moderately priced DSP chip. We must also fully consider the after-sales service and other factors provided by the manufacturer. Good after-sales technical support is also an important resource in the development process.
(8) Other factors: including the package form of the DSP chip, environmental requirements, delivery cycle, life cycle, etc.
Broadly speaking, digital signal processing is a technical discipline that studies the analysis, transformation, filtering, detection, modulation, demodulation, and fast algorithms of signals using digital methods. However, many people believe that digital signal processing is mainly about digital filtering techniques, discrete transform fast algorithms and spectral analysis methods. With the development of digital circuits and system technology and computer technology, digital signal processing technology has also been developed accordingly, and its application fields are very extensive.
Digital control and motion control applications include disk drive control, engine control, laser printer control, inkjet printer control, motor control, power system control, robot control, high-precision servo system control, and CNC machine tools.
Applications for low-power, handheld devices, and wireless terminals include: mobile phones, PDAs, GPS, digital radio stations, and so on.
Digital filter
There are many practical types of digital filters, which can be divided into two types: finite impulse response type and infinite impulse response type. They can be implemented in both hardware and software. In the hardware implementation, it consists of an adder, a multiplier, etc., which is completely different from the analog filter formed by resistors, inductors, and capacitors. The digital signal processing system is easy to use with digital integrated circuits, showing the advantages of small size, high stability, and programmable. Digital filters can also be implemented in software. The software implementation method is to perform a digital filtering calculation by means of a general-purpose digital computer according to a filter design algorithm.
2. Fourier transform
In 1965, JW Curry and TW Tuji first proposed a fast algorithm of discrete Fourier transform, referred to as fast Fourier transform, which is represented by FFT. Since the fast algorithm, the number of operations of the discrete Fourier transform is greatly reduced, making the realization of digital signal processing possible. The Fast Fourier Transform can also be used to perform a series of related fast operations, such as correlation, convolution, power spectrum, and so on. The Fast Fourier Transform can be made into a dedicated device or implemented in software. Similar to the fast Fourier transform, other forms of transforms, such as Walsh transforms, number theory transforms, etc., can also have their own fast algorithms.
3. Spectral analysis
An analysis method that describes the characteristics of a signal in the frequency domain can be used not only for deterministic signals but also for random signals. The so-called deterministic signal can be represented by a given time function, which is deterministic at any time; the random signal does not have such a characteristic that its value at a certain moment is random. Therefore, random signal processing can only be based on stochastic process theory, using statistical methods for analysis and processing, such as the use of mean, mean square, variance, correlation function, power spectral density function and other statistics to describe the characteristics of random processes or random The characteristics of the signal.
Speech Signal Processing Speech signal processing is one of the important branches in signal processing. The main aspects it includes include: speech recognition, language understanding, speech synthesis, speech enhancement, speech data compression, and more. Various applications have their own special problems. The speech recognition is to extract the feature parameters of the speech signal to be recognized instantaneously, and match with the known speech samples, thereby determining the phoneme attribute of the speech signal to be recognized. With regard to speech recognition methods, there are statistical pattern speech recognition, structure and sentence pattern speech recognition. Using these methods, important parameters such as formant frequency, pitch, voice, noise, etc. can be obtained. Speech understanding is the theory and technique of dialogue between human and computer using natural language. basis. The main purpose of speech synthesis is to enable computers to speak. To this end, it is first necessary to study the variation of the speech characteristic parameters with time in the pronunciation, and then use the appropriate method to simulate the process of pronunciation, and synthesize it into language. Other related language processing issues also have their own characteristics. Speech signal processing is the basis for the development of intelligent computers and intelligent robots, and is the basis for the manufacture of vocoders. Speech signal processing is a rapidly developing signal processing technology.
4. Image signal processing
The application of image signal processing has penetrated into various fields of science and technology. For example, image processing techniques can be used to study the trajectories of particles, the structure of biological cells, the state of geomorphology, the analysis of meteorological clouds, and the composition of cosmic stars. In the practical application of image processing, remote sensing image processing technology, tomographic imaging technology, computer vision technology and scene analysis technology have been obtained. According to the application characteristics of image signal processing, processing techniques can be roughly divided into image enhancement, recovery, segmentation, recognition, coding and reconstruction. These processing technologies have their own characteristics and are rapidly evolving.
5. Vibration signal processing
The analysis and processing technology of mechanical vibration signals has been applied in the research and production of automobiles, airplanes, ships, machinery and equipment, building construction, dam design and so on. The basic principle of vibration signal processing is to add an exciting force to the test body as an input signal. The output signal is monitored at the measurement point. The ratio of the output signal to the input signal is called the transfer function (or transfer function) of the system composed of the test body. The so-called modal parameter identification is performed according to the obtained transfer function, and the main parameters such as modal stiffness and modal damping of the system are calculated. This establishes a mathematical model of the system. In turn, a dynamic optimization design of the structure can be made. These tasks can all be performed using a digital processor. This type of analysis and processing is generally referred to as modal analysis. In essence, it is a special method used in signal processing in vibration engineering.
6. Geophysical processing
In order to explore oil and gas and other mineral deposits stored deep underground, seismic exploration methods are often used to detect stratigraphic structure and lithology. The basic principle of this method is to apply artificial shocks at a selected location, such as using a method of explosion to generate a vibration wave to propagate underground. When the boundary layer is formed, a reflected wave is generated, and a column is placed far away from the vibration source. A susceptor that receives reflected waves that reach the ground. The depth and structure of the formation are judged from the delay time and intensity of the reflected wave. The seismic records received by the susceptors are complex and require processing for geological interpretation. There are many methods of processing, such as deconvolution, homomorphic filtering, etc., which is an issue that is still being studied.
7. Biomedical treatment
Signal processing is mainly used in biomedical research to assist in the research of basic theories of biomedicine and for diagnostic examination and monitoring. For example, basic theoretical research in cytology, brain neurology, cardiovascular science, genetics, etc. The human cranial nervous system is made up of about 10 billion nerve cells and is a very complicated and huge information processing system. In this processing system, the transmission and processing of information are performed in parallel, and have special functions. Even if a certain part of the system is in trouble, other parts can still work, which is impossible for the computer. Therefore, the research on the information processing model of the human brain has become an important topic in basic theoretical research. In addition, the study of neural cell models, the study of chromosomal functions, etc. can be carried out by means of the principles and techniques of signal processing.
V. Future DSP development trendFor a long period of time, wireless applications will still be the driving engine for the programmable DSP market. DSP technology will continue to be the primary means of accessing narrowband, broadband or wireless Internet. It is also a key technology in the emerging packet (IP) telephony market, and DSP will continue to be the technology driver for the entire semiconductor industry.
In product applications, currently important DSP applications, such as mobile phones, modems, HDD and other personal computer and communications applications. But there are indications that the digital revolution has promoted the widespread use of high-performance DSPs such as digital cameras, VoIP phones and handheld electronic devices. China's development in consumer electronics is even more compelling. Many consumer electronics products require lower-cost, easier-to-use DSP products such as audio equipment, DVRs, and set-top boxes. Overall, DSP applications in the field of communications, digital audio and video products will become more and more popular.
Electric Test Pen ,Battery Detector,Circuit Tester,Voltage Test Pen
YINTE TOOLS (NINGBO) CO., LTD , https://www.yinte-tools.com