What Are the Different Types of Mergers?

The combining method refers to that after receiving the M signals, the receiving end uses certain technical means to add the input M independent fading signals and then combines and outputs to reduce the influence of fading. Commonly used selective merging.

The combining method refers to that after receiving the M signals, the receiving end uses certain technical means to add the input M independent fading signals and then combines and outputs to reduce the influence of fading. Commonly used selective merging.

Chinese name

Merger method

Foreign name

Combining Method

Applied discipline

Communication

definition

The merge method is usually applied to spatial diversity. After receiving the M independent branch signals at the receiving end, the diversity gain can be obtained by combining techniques. According to the location of the merge technology used at the receiving end, it can be divided into pre-detection merge technology and post-detection merge technology.

classification

Selective combining is to detect the signals of all diversity branches to select the one with the highest signal-to-noise ratio as the output of the cooperator, which is also called switching addition.

Assume that the voltages of the M input signals are r1 (t), r2 (t), ..., rM (t), and the combined output is r (t). R (t) can be expressed as:

In the formula, the weighting coefficient of the k-th signal. Selecting different weighting factors can constitute different ways of combining. At present, there are four main types of merger: selective merger, feedback merger, maximum ratio merger, and equal gain merger.

2.1 Choose to merge

In the selective combiner, there is only one effective weighting factor, and the rest are all zero. FIG. 1 is a schematic diagram of selective combining. The intermediate frequency signals of the two branches are respectively modulated, and then the signal-to-noise ratio is compared. The branch with the higher signal-to-noise ratio is selected and connected to the output portion of the receiver.

The selection of the combining method is simple and easy to implement, but because the unselected branch signals are discarded, the anti-fading performance is not optimal. However, if the combination is implemented at intermediate or high frequencies, the signals of the branches must be in phase, which will increase The complexity of the circuit.

2.2 Feedback or scan merge

Scanning and merging are very similar to selecting and merging, but instead of always using the best of the M branches, it scans the M branches in a fixed order until it finds that the signal of a certain branch exceeds a preset threshold Then send this signal to the receiver. Once the signal is below the threshold, the scanning process will restart. Its working principle is shown in Figure 2.

Figure 2 Schematic of feedback scan merge

2.3 Maximum Ratio Merger

The maximum ratio combining method is a better combining method. It weights M signals, and the weight is determined by the ratio of the signal voltage and noise power corresponding to each signal. Its merging principle is shown in Figure 3.

The signal envelope rk (t) of each branch is represented by rk, and the weighting coefficient ak of each branch is proportional to the signal envelope rk and inversely proportional to the noise power Nk, that is:

It can be known from this that the output signal envelope of the maximum combining ratio combiner is:

2.4 equal gain combining

In some cases, it is inconvenient to generate variable weights by the need to combine according to the maximum ratio, so equal gain combining occurs. In this method, the signals of the branches are added in phase and then added. When added, the weights of the branches are the same. In this way, the receiver can still use the signals received at the same time, and the probability that the receiver can synthesize a demodulated signal from a large number of signals that cannot be demodulated is still very large.

The equal gain combining method is relatively easy to implement, and its performance is close to the maximum ratio combining. This is because the equal gain combining does not need to weight the signals, and the signals of the branches are added with equal gain.

The principle of equal gain combining is shown in Figure 4.

It can be seen that among the three merge methods, the performance of the maximum ratio merge is the best, and the performance of the selective merge is the worst. When N is large, the combined gain of equal gain combining is close to the combining gain of maximum ratio combining.