What are the purposes and types of source/channel coding?
This picture is a picture that everyone will encounter when learning communication principles. It contains three elements: information source, channel and information sink. This picture can directly answer the original question. The so-called source coding is for source coding, and the so-called channel coding is for channel coding.
Effectiveness : In order to reduce redundancy, Huffman coding is commonly used. This Huffman code is even mentioned in computer courses. In fact, its essential uses are the same, but the scenarios are different.
Reliability : In order to reduce errors in signal transmission, typical error correction codes, turbo codes, convolutional codes, etc. are commonly used.
What are the requirements of Shannon's formula on the channel and input signal? (prerequisite)
You might as well take a look at the proof of Shannon's formula in the classic textbook "Modern Communication Principles and Technology" by Ms. Cao Lina
We can get several prerequisites:
- The first is the discrete channel
- Both the input signal and noise are Gaussian distributed, ensuring maximum mutual information in discrete systems.
- In the derivation, the sampling rate is required to be 2 B 2B2B _
The necessary and sufficient conditions for linear phase FIR filters published in "Education and Teaching Research" , an article written by Professor Wang Shikui of Chongqing Three Gorges University, is of great reference significance. Students who are interested can read it.
What is the Gibbs effect?
Gibbs phenomenon (also called Gibbs effect): After Fourier series expansion of periodic functions with discontinuous points (such as rectangular pulses), finite terms are selected for synthesis. When more terms are selected, the peaks appearing in the synthesized waveform will be closer to the discontinuity point of the original signal. When the number of selected items is large, the peak value tends to a constant, which is approximately equal to 9% of the total jump value.
Xie Huimin’s Mathematical Analysis Exercise Course Handouts Volume 2
The Pelliwiener criterion is a physically achievable derivation of the filter. Its main purpose is to make the impulse response of the filter causal.
What is inter-symbol interference?
-
Intra-symbol interference: OFDM symbols contain many frequencies. Due to the frequency selectivity of the channel, different frequency components in the same symbol have different signal attenuation when they reach the receiving end, so errors will occur.
-
ISI: Inter-symbol interference (Symbol Interference). The ISI here and the inter- code crosstalk we talked about before are two different concepts. For a certain subcarrier, the ISI mentioned in the OFDM system refers specifically to the iiThe impact of the i path signal on the first path.
Methods to eliminate multipath interference in OFDM
ISI: OFDM solves ISI through guard intervals . This guard interval can be filled with ZP (zero padding) or CP (cyclic prefix) or CS (cyclic suffix).
The difference between diversity and multiplexing
- Diversity: The same data is transmitted on multiple independent paths. The receiving end uses diversity combining technology to resist channel fading, improve transmission reliability, and reduce bit error rates.
- Multiplexing: Transmitting different data on multiple independent sub-channels on the same path, making full use of system resources and increasing system capacity, that is, the total data rate.
Classification of diversity technologies
Macrodiversity is mainly used in cellular wireless systems
- Mainly used in diversity technology for cellular systems. In macrodiversity, multiple base stations are set up in different geographical locations and in different directions to communicate with a mobile station in the community at the same time. As long as the signal propagation in all directions is not affected by shadow effects or terrain at the same time and suffers from severe slow fading, this method can ensure that the communication will not be interrupted. It is a technique for reducing slow fading.
Microdiversity is a diversity technique that reduces the effects of fast fading
-
The basic principle of space diversity is to receive the same signal at any two different locations. As long as the distance between the two locations is large enough, the fading of the signals received at the two locations is irrelevant.
-
The basic principle of frequency diversity is that the fading of two signals whose frequency separation is greater than the correlation bandwidth is uncorrelated . Therefore, the same information can be transmitted using multiple frequencies to achieve frequency diversity.
-
Time diversity: In addition to spatial and frequency independence, fast fading also has time independence, that is, the same signal is retransmitted multiple times at different times and intervals. As long as the time interval between each transmission is large enough, then the appearance of each transmitted signal will be The declines of will be independent of each other
Shannon formula
- Shannon's first theorem (variable length distortion-free source coding theorem)
- Shannon's second theorem (noisy channel capacity theorem)
- Shannon's third theorem (distorted source coding theorem under the distortion-preserving criterion)
5G key technologies
- millimeter wave
- Massive MIMO
- UDN, miniaturization of base stations, densification of cells, diversification of nodes and high degree of collaboration, Ultra-Dense Network
5G application scenarios
The above figure shows the three major scenarios of 5G, namely ultra-high-definition video bandwidth, large-scale IoT services, and driverless/automated low-latency services.
Three major visions of 6G
-
Terahertz communication: Terahertz ultra-high-speed wireless network is a new type of wireless network. Different from traditional wireless networks, it operates in the terahertz frequency band and can support data transmission rates of several 10Gbps or even 1Tbps. The frequency band of the terahertz band is relatively wide and most of it has not yet been allocated for use. It can carry Gbps data volume.
The attenuation characteristics of terahertz make terahertz communication more difficult over a distance of several meters, making terahertz communication more suitable for short-distance communication scenarios. -
Communication perception integration: On the one hand, the entire communication network can be used as a huge sensor. Network elements send and receive wireless signals. Using the transmission, reflection and scattering of radio waves, the physical world can be better perceived and understood.
On the other hand, the high-precision positioning, imaging and environment reconstruction capabilities provided by sensing can help improve communication performance, such as more accurate beamforming, faster beam failure recovery, and terminal channel state information (Channel State Information, CSI) tracking. Lower overhead. -
Intelligent metasurface technology: RIS is an emerging technology that can define new wireless transmission and propagation modes and control communication channels (reshape wireless channels). In other words, a RIS is a metasurface containing electronically controllable and low-power analog processing elements. The absorption, reflection, refraction and phase of passive reflective elements can be adjusted in real time to guide incident electromagnetic signals in the desired direction.
Coherence time and coherence bandwidth
-
Coherence bandwidth: Coherence bandwidth describes the delay spread . Coherence bandwidth is an important parameter characterizing the characteristics of multipath channels . It refers to a specific frequency range in which any two frequency components have strong Magnitude correlation, that is, the multipath channel has constant gain and linear phase within the coherence bandwidth. Typically, the coherence bandwidth is approximately equal to the reciprocal of the maximum multipath delay . From a frequency domain perspective, if the coherence bandwidth is smaller than the bandwidth of the transmitting channel, the channel characteristics will cause frequency selective fading of the received signal waveform, that is, the amplitude of some frequency component signals can be enhanced, while the amplitude of other frequency component signals can be enhanced. will be weakened.
-
Coherence time: describes the Doppler spread, and coherence time describes the time-varying characteristics of the channel in the time domain. Coherence time is inversely proportional to Doppler spread and is the statistical average of the time intervals during which the channel impulse response remains constant . If the symbol period of the baseband signal is greater than the coherence time of the channel, the channel may change during the transmission of the baseband signal, causing distortion of the received signal, resulting in time-selective fading, also known as fast fading; if the symbol period of the baseband signal is smaller than the channel coherence time, the channel will not change during the transmission of the baseband signal, and time-selective fading will not occur, also known as slow fading.
Coherence time is the maximum time difference range within which the channel remains constant, and coherence bandwidth is similar, which is the maximum frequency difference range within which the channel remains constant. It is easier to understand this concept from the perspective of diversity: Time diversity requires that the time of two transmissions be greater than the coherence time of the channel, that is, if the transmission time is less than the coherence time of the channel, the signals transmitted twice will experience the same fading, and diversity resistance The role of fading does not exist, and the coherence bandwidth can be understood from frequency diversity.
Classification of wireless channel weakness
When the signal encounters obstacles, the signal will change randomly, and the surrounding environment at different locations is very different, resulting in random changes in signal power at a given distance. The lognormal shading model is most commonly used to characterize this random decay.
Fading model for multipath channels
-
Rayleigh channel: channel fading model that takes into account multipath effects, but does not have a major direct component
-
Rician Channel: When there is a dominant (non-debilitating) channel component, such as line-of-sight propagation.
-
Clarke and Nakagami statistical channel model
Key technologies in 3G and 4G
The core technologies in 4G mobile communications include MIMO, OFDM, coverage area switching technology and software radio technology.
Software Radio Technology SDR technology: refers to technology that can use software to program, control, and specify radios. Adaptable to multiple standards such as Bluetooth, WiFi, GNSS, 4G and 5G.
The core technology of 3G is CDMA technology, which can realize signal spreading. Code Division Multiple Access.
Summary of FIR and IIR filters
FIR filter design methods: window function method, frequency sampling method, optimization method (least squares method)
Design methods of IIR filters: impulse response invariant method, step response invariant method, bilinear change method
Stability comparison:
- IIR filter: The pole of the transfer function can be located anywhere within the unit circle, which has stability problems
- FIR filter: The pole of the transfer function is fixed at the origin, and there is no stability problem.
Performance comparison:
- IIR filter: high efficiency (efficiency is reflected in: reaching the target at a lower order) at the expense of phase nonlinearity. The better the selectivity, the more severe the phase nonlinearity. If the IIR filter wants to obtain a linear phase and meet the technical requirements of amplitude filtering, an all-pass network must be added for phase correction, which will increase the order of the filter.
- FIR filter: Strict linear phase can be obtained
Compare algorithmically:
- IIR filter: no fast algorithm
- FIR filter: The impulse response is finitely long, and the FFT algorithm can be used to reduce the amount of calculations.
Compare structurally:
- IIR filter: a recursive structure must be used, and the pole position must be within the unit circle. In recursive structures, operations are rounded, sometimes causing parasitic oscillations
- FIR filter: mainly adopts non-recursive structure, there is no stability problem, and the operation error is also small