CN103236867B - Radiofrequency filter switching system for GPIB (general purpose interface bus) command control - Google Patents

Abstract

A radiofrequency filter switching system for GPIB (general purpose interface bus) command control comprises a central processing unit TMS320F2808, a GPIB, a two-level coaxial switch, a first filter, a second filter, a third filter, a fourth filter, a fifth filter and a sixth filter. After detected signals pass the filters, on and off of the coaxial switch are controlled to filter the detected signals through the GPIB according to GPIB commands designed in the central processing unit TMS320F2808.

Description

The radio-frequency filter switched system that GPIB order controls

Technical field

The invention belongs to electronic technology/communications field, relate to a kind of radio-frequency filter switched system utilizing central processing unit TMS320F2808 to be controlled by GPIB order, for checking and approving the filtering of tested signal in radiation disturbance and radiation dispersion test macro, be specially adapted to check and approve building of the test of radiation dispersion wireless terminal radio-frequency (RF) index and integrated test system.

Background technology

Conventional terminal radio frequency Auto-Test System refers to employing computer control, automatically complete and set up call, link switching, signal measurement, data computing the automatization test system outputed test result, be mainly used in wireless terminal radio-frequency (RF) index test and integrated test system build, comprise the wireless terminal such as TD-SCDMA, GSM, WLAN, WCDMA, CDMA and bluetooth radio-frequency (RF) index test and Auto-Test System build;

The radio frequency test system of prior art is primarily of external minority manufacturer production, production cycle is long, and limited degree is large, expensive, and switched system can not manually control separately, be difficult to be applied to different test environments flexibly, present stage terminal radio frequency test is main uses advanced test instrumentation, but carry out in test process at radio frequency terminal equipment, for different test events, need to carry corresponding filtering radio circuit to meet test request, simultaneously need many filtering radio frequency links to carry completing in a radio frequency testing process;

Existing filtering radio frequency test system mainly uses advanced test instrumentation, but carry out in test process at radio frequency terminal equipment, for different test events and standard, need to carry corresponding radio frequency link to meet test request, simultaneously need many radio circuits to carry completing in a radio frequency testing process.If manually carry test link, can measure error be introduced, affect the accuracy of test result, and the filter effect of prior art is not good.

Summary of the invention

Technology of the present invention is dealt with problems and is: the deficiency overcoming existing control technology, provides a kind of radio-frequency filter switched system for checking and approving EMC emission test based on central processing unit, gpib bus, coaxial switch and filter composition.

Technical solution of the present invention is: the radio-frequency filter switched system that a kind of GPIB order controls, be made up of central processing unit TMS320F2808, gpib bus, 2 grades of coaxial switches, the first filter, the second filter, the 3rd filter, the 4th filter, the 5th filter and the 6th filters, wherein, microprocessor TMS320F2808 is by gpib bus for controlling turning on and off of 2 grades of coaxial switches, and tested signal is by exporting the signal of specifying after 6 path filters;

Described first filter is 1.6GHz high pass filter, mainly completes and carries out filtering to below 1.60GHz signal, retains more than 1.60GHz signal and passes through test link smoothly; Second filter is 2.4GHz high pass filter, mainly completes and carries out filtering to below 2.40GHz signal, retains more than 2.40GHz signal and passes through test link smoothly; 3rd filter is 3.2GHz high pass filter, mainly completes and carries out filtering to below 3.20GHz signal, retains more than 3.20GHz signal and passes through test link smoothly; 4th filter is 800MHz low pass filter, mainly completes and carries out filtering to more than 800MHz signal, retains below 800MHz signal and passes through test link smoothly; 5th filter is 600MHz low pass filter, mainly completes and carries out filtering to more than 600MHz signal, retains below 600MHz signal and passes through test link smoothly; 6th filter is 400MHz low pass filter, mainly completes and carries out filtering to more than 400MHz signal, retains below 400MHz signal and passes through test link smoothly.

Described filter at different levels is grouped into by high-pass part and band resistance part.

The present invention's advantage is compared with prior art:

The radio-frequency filter switched system controlled based on digital signal processor TMS320F2808 of the present invention, solve the problems such as prior art equipment de-sign poor controllability, integrated poor performance, achieve the automatic switchover of radio frequency link, eliminate interference because artificially introducing or leakage and increase test error; The filtering mode that simultaneously the present invention combines owing to have employed high-pass filtering and bandreject filtering, had the advantage of conventional iir filter and FIR filter concurrently, time delay is little, precision is high, improves the real-time of radio-frequency filter switched system.

Accompanying drawing explanation

Fig. 1 for the present invention is based on central processing unit TMS320F2808, GPIB order control radio-frequency filter switched system.

Fig. 2 is central processing unit TMS320F2808 circuit diagram of the present invention.

Embodiment

The radio-frequency filter switched system that a kind of GPIB order controls, as shown in Figure 1, be made up of central processing unit TMS320F2808, gpib bus, 2 grades of coaxial switches, the first filter, the second filter, the 3rd filter, the 4th filter, the 5th filter and the 6th filters, wherein, microprocessor TMS320F2808 is by gpib bus for controlling turning on and off of 2 grades of coaxial switches, and tested signal is by exporting the signal of specifying after 6 path filters;

Described first filter is 1.6GHz high pass filter, mainly completes and carries out filtering to below 1.60GHz signal, retains more than 1.60GHz signal and passes through test link smoothly; Second filter is 2.4GHz high pass filter, mainly completes and carries out filtering to below 2.40GHz signal, retains more than 2.40GHz signal and passes through test link smoothly; 3rd filter is 3.2GHz high pass filter, mainly completes and carries out filtering to below 3.20GHz signal, retains more than 3.20GHz signal and passes through test link smoothly; 4th filter is 800MHz low pass filter, mainly completes and carries out filtering to more than 800MHz signal, retains below 800MHz signal and passes through test link smoothly; 5th filter is 600MHz low pass filter, mainly completes and carries out filtering to more than 600MHz signal, retains below 600MHz signal and passes through test link smoothly; 6th filter is 400MHz low pass filter, mainly completes and carries out filtering to more than 400MHz signal, retains below 400MHz signal and passes through test link smoothly.

Described filter at different levels is grouped into by high-pass part and band resistance part.

Fig. 2 is central processing unit TMS320F2808 circuit diagram of the present invention, due to TMS320F2808 have abundant in establish module and interrupt resources thereof, so programming flexibility is very strong.The programming of this application system takes full advantage of the abundant interrupt resources of DSP, so main program is very simple.Mainly comprise three modules: system initialization module, inside and outsidely establish unit initialization module, software initialization module.

1, dsp system initialization module programming

The function of system initialization carries out initialization to the macrosystem of DSP:

1. set memory space to arrange, define a certain interval storage program;

2. system clock is arranged, and system clock frequency is set to 150MHz;

3. system house dog is arranged, not enable house dog;

4. system break is arranged, and open system is always interrupted, clear all interrupt identifications, enable 1 grade of interrupt INT 1,3 grades of interrupt INT 3,5 grades of interrupt INT 5;

2, inside and outside initialization of establishing unit

After dsp system initialization, configure corresponding register, universal input/output (GPIO) multiplexer, A/D module sampling period and PWM carrier cycle etc. are set.

TMS320F2808 chip has nearly 56 multifunctional pin, and they using the I/O pin of these pins as peripheral hardware in sheet, when not using peripheral hardware in sheet, also can be used as digital I/O mouth by user.The GPIO multiplexer of TMS320F2808, when relevant pin is used as digital I/O mouth, can form digital I/O mouth GPIOA, GPIOB of two 16; The I/O mouth GPIOD of one 4; The digital I/O mouth GPIOF of I/O mouth GPIOE and 15 of one 3.Use the relevant register of GPIO can select and control the operation of these shared pins.As: GPxMUX function mask register can be used to configure I/O and to be operated in peripheral hardware operator scheme or digital quantity I/O pattern.The direction of GPxDIR register configuration I/O can be used.In this application system, GPIOA1, GPIOA2, GPIOA3, GPIOA4, GPIOA5, GPIOA6 are configured to peripheral functionality, produce interrupt signal, coaxial switch being attached thereto, for controlling the break-make of coaxial switch and major loop.

3, software initialization module installation

During this part initialization mainly runs program, used variable register and temporary register compose initial value, such as external interrupt XINT1 flag bit, seizure interrupt flag bit etc.

2 grades of radio-frequency (RF) coaxial switching circuits of the present invention are formed by connecting primarily of AQW212EH, resistance (R360).Send signal by TMS320F2808, be responsible for by GPIB order the break-make controlling coaxial switch.This part power supply part is provided by+24V, and another part provides low and high level by the general GPIO pin of TMS320F2808.

The method for designing step of each filter of 6 filters of the present invention is as follows:

(1) according to the requirement of output signal, determine that the parameter with high-pass part is divided in the band resistance part of filter, comprising: stopband attenuation COEFFICIENT K is divided in band resistance part_decay, centre frequency K_f0; High-pass part ripple coefficient K_pass, high-pass part time delay coefficient of variation K_delay; Filter cell total stopband attenuation coefficient A_decay, filter peak frequency K_fmaxand intensity A_s, wherein filter peak frequency K_fmaxand intensity A_semploying power spectrum analysis method obtains: stopband attenuation parameter K is divided in band resistance part_decay=0.6A_s, centre frequency K_f0=K_fmax; High-pass part passband ripple COEFFICIENT K_pass=0.02dB, high-pass part time delay coefficient of variation K_delay=1.2; Filter cell total stopband attenuation coefficient A_decaymore than 10 times that signal strength signal intensity in passband exceedes noise intensity in stopband should be met;

(2) with centre frequency K_f0=K_fmax, the band resistance part of designing filter is divided, and obtains transfer functionwherein numerator coefficients is a_s1, a_s2..., a_sn, denominator coefficients is b_s0, b_s1, b_s2..., b_sn.

(3) in time domain, Least Square Method is utilized to make G_hcz () phase linearity, obtains G_hc(z) initial molecular coefficient a₁(0), a₂(0) ..., a_nand denominator coefficients b (0)₀(0), b₁(0) ... b_n(0), namely the transfer function of initial high-pass filtering part is$G_{\mathrm{hc}}\left(z\right)=\frac{b_0\left(0\right)+b_1\left(0\right)z^{-1}+b_2\left(0\right)z^{-2}+...+b_n\left(0\right)z^{-n}}{1+a_1\left(0\right)z^{-1}+a_2\left(0\right)z^{-2}+...+a_n\left(0\right)z^{-n}};$

Least Square Method is utilized to make G in this step_hcz () phase linearity, obtains G_hc(z) initial coefficients a₁(0), a₂(0) ..., a_n(0), b₀(0), b₁(0) ... b_n(0) step is as follows:

A () is with coefficient a to be asked₁, a₂..., a_n, b₀, b₁b_nset up G_hc(z) model,$G_{\mathrm{hc}}\left(z\right)=\frac{b_0+b_1{z}^{-1}{+b}_2{z}^{-2}+...+b_n{z}^{-n}}{1+a_1{z}^{-1}+a_2{z}^{-2}+...+a_n{z}^{-n}};$The high-pass part of design linear phase filter$G_f\left(z\right)=\underset{k=0}{\overset{N-1}{\mathrm{\Σ}}}{h}_k{z}^{-k},$Its exponent number is G_hcz the twice of () exponent number, passband ripple compares G_hc(z) large 10 times, stopband attenuation ratio G to be asked_hcz () is little 10 times;

B () sets up least-squares estimation equation.With a₁, a₂..., a_n, b₀, b₁, b₂b_nestimated state X (k)=[a of composition least-squares estimation₁, a₂..., a_n, b₀, b₁..., b_n]^t, with G_hcz () input x (k) and output y (k) form least squares estimator and survey matrix G (k)=[-y (k-1),-y (k-n), x (k), x (k-n)], set up least-squares estimation equation:

Z(k)＝G(k)X(k)+B(k)

Wherein Z (k) is k moment measurement amount, and X (k) is k moment system state amount, and B (k) is k moment measurement noise matrix, k=1...n;

Adopt the Least Square Recurrence method of the uncertainty factor, algorithm is as follows:

$\hat{X}\left(k\right)=\hat{X}(k-1)+K\left(k\right)[Z\left(k\right)-G\left(k\right)\hat{X}(k-1)]$

K(k)＝C(k-1)G^T(k)[G(k)C(k-1)G^T(k)+u(k)]^-1

$C\left(k\right)=\frac{[E-K\left(k\right)G\left(k\right)]C(k-1)}{u\left(k\right)}$

Whereinfor k moment system state estimation amount,for k-1 moment system state estimation amount, K (k) is k moment gain matrix, Z (k) is k moment measurement amount, C (k) is the covariance matrix of k moment system estimation amount, C (k-1) is the covariance matrix of k-1 moment system estimation amount, u (k) is the k moment uncertainty factor, and 0.7 < u (k) < 0.9, E is unit matrix;

C () utilizes least-squares algorithm in step (b) under white Gaussian noise excitation, get least-squares estimation quantity of state initial valuecovariance of estimator matrix initial value C (0)=constE, const > 0 is constant, makes G_hcz () approaches the high pass filter G in (a)_fz (), to obtain the G with linear phase_hcz (), its coefficient is a₁(0), a₂(0) ..., a_n(0), b₀(0), b₁(0) ... b_n(0).

(4) by G (z)=G_bs(z) G_hcz () asks for total transfer function G (z) of filter, and gather filtered signal and carry out power spectrumanalysis, gathers 1s filtered signal carry out power spectrumanalysis, if noise attentuation is lower than A in G (z) stopband with 2KHz_decay, increase K according to the relations of 2 times_decay, i.e. K_decay=2K_decay, re-start step (2)-step (4), try to achieve the numerator coefficients b of G (z)₀(n), b₁(n), b₂(n) ... b_n(n) and denominator coefficients a₁(n), a₂(n) ..., a_nn (), obtains the transfer function of filter$G\left(z\right)=\frac{b_0\left(n\right)+b_1\left(n\right)z^{-1}+b_2\left(n\right)z^{-2}+...+b_n\left(n\right)z^{-n}}{1+a_1\left(n\right)z^{-1}+a_2\left(n\right)z^{-2}+...+a_n\left(n\right)z^{-n}}$

The content be not described in detail in specification of the present invention belongs to the known prior art of professional and technical personnel in the field.

Claims (1)

1. the radio-frequency filter switched system of a GPIB order control, it is characterized in that: by central processing unit TMS320F2808, gpib bus, 2 grades of coaxial switches, first filter, second filter, 3rd filter, 4th filter, 5th filter and the 6th filter composition, described first to the 6th filter is positioned between 2 grades of coaxial switches side by side, complete filter by coaxial switch to switch, wherein, microprocessor TMS320F2808 by gpib bus for controlling turning on and off of 2 grades of coaxial switches, tested signal is by exporting the signal of specifying after 6 path filters, TMS320F2808 can form digital I/O mouth GPIOA, GPIOB of two 16, the I/O mouth GPIOD of one 4, the digital I/O mouth GPIOF of I/O mouth GPIOE and 15 of one 3, GPxMUX function mask register configuration I/O is used to be operated in peripheral hardware operator scheme or digital quantity I/O pattern, use the direction of GPxDIR register configuration I/O, within the system, GPIOA1, GPIOA2, GPIOA3, GPIOA4, GPIOA5, GPIOA6 are configured to peripheral functionality, produce interrupt signal, coaxial switch is attached thereto, for controlling the break-make of coaxial switch and major loop,

Described first filter is 1.6GHz high pass filter, mainly completes and carries out filtering to below 1.60GHz signal, retains more than 1.60GHz signal and passes through smoothly; Second filter is 2.4GHz high pass filter, mainly completes and carries out filtering to below 2.40GHz signal, retains more than 2.40GHz signal and passes through smoothly; 3rd filter is 3.2GHz high pass filter, mainly completes and carries out filtering to below 3.20GHz signal, retains more than 3.20GHz signal and passes through smoothly; 4th filter is 800MHz low pass filter, mainly completes and carries out filtering to more than 800MHz signal, retains below 800MHz signal and passes through smoothly; 5th filter is 600MHz low pass filter, mainly completes and carries out filtering to more than 600MHz signal, retains below 600MHz signal and passes through smoothly; 6th filter is 400MHz low pass filter, mainly completes and carries out filtering to more than 400MHz signal, retains below 400MHz signal and passes through smoothly;

Described filter at different levels is grouped into by high-pass part and band resistance part;

The method for designing step of each filter of 6 filters adopted is as follows:

(1) according to the requirement of output signal, determine that the parameter with high-pass part is divided in the band resistance part of filter, comprising: stopband attenuation COEFFICIENT K is divided in band resistance part_decay, centre frequency K_f0; High-pass part ripple coefficient K_pass, high-pass part time delay coefficient of variation K_delay; Filter cell total stopband attenuation coefficient A_decay, filter peak frequency K_fmaxand intensity A_s, wherein filter peak frequency K_fmaxand intensity A_semploying power spectrum analysis method obtains: stopband attenuation parameter K is divided in band resistance part_decay=0.6A_s, centre frequency K_f0=K_fmax; High-pass part passband ripple COEFFICIENT K_pass=0.02dB, high-pass part time delay coefficient of variation K_delay=1.2; Filter cell total stopband attenuation coefficient A_decaymore than 10 times that signal strength signal intensity in passband exceedes noise intensity in stopband should be met;

(2) with centre frequency K_f0=K_fmax, the band resistance part of designing filter is divided, and obtains transfer functionwherein numerator coefficients is a_s1, a_s2..., a_sn, denominator coefficients is b_s0, b_s1, b_s2..., b_sn;

(3) in time domain, Least Square Method is utilized to make G_hcz () phase linearity, obtains G_hc(z) initial molecular coefficient a₁(0), a₂(0) ..., a_nand denominator coefficients b (0)₀(0), b₁(0) ... b_n(0), namely the transfer function of initial high-pass filtering part is

Least Square Method is utilized to make G in this step_hcz () phase linearity, obtains G_hc(z) initial coefficients a₁(0), a₂(0) ..., a_n(0), b₀(0), b₁(0) ... b_n(0) step is as follows:

A () is with coefficient a to be asked₁, a₂..., a_n, b₀, b₁b_nset up G_hc(z) model,the high-pass part of design linear phase filterits exponent number is G_hcz the twice of () exponent number, passband ripple compares G_hc(z) large 10 times, stopband attenuation ratio G to be asked_hcz () is little 10 times;

B () sets up least-squares estimation equation, with a₁, a₂..., a_n, b₀, b₁, b₂b_nestimated state X (k)=[a of composition least-squares estimation₁, a₂..., a_n, b₀, b₁..., b_n]^t, with G_hcz () input x (k) and output y (k) form least squares estimator and survey matrix G (k)=[-y (k-1),-y (k-n), x (k), x (k-n)], set up least-squares estimation equation:

Z(k)＝G(k)X(k)+B(k)

Wherein Z (k) is k moment measurement amount, and X (k) is k moment system state amount, and B (k) is k moment measurement noise matrix, k=1...n;

Adopt the Least Square Recurrence method of the uncertainty factor, algorithm is as follows:

K(k)＝C(k-1)G^T(k)[G(k)C(k-1)G^T(k)+u(k)]^-1

Whereinfor k moment system state estimation amount,for k-1 moment system state estimation amount, K (k) is k moment gain matrix, Z (k) is k moment measurement amount, C (k) is the covariance matrix of k moment system estimation amount, C (k-1) is the covariance matrix of k-1 moment system estimation amount, u (k) is the k moment uncertainty factor, and 0.7 < u (k) < 0.9, E is unit matrix;

C () utilizes least-squares algorithm in step (b) under white Gaussian noise excitation, get least-squares estimation quantity of state initial valuecovariance of estimator matrix initial value C (0)=constE, const > 0 is constant, makes G_hcz () approaches the high pass filter G in (a)_fz (), to obtain the G with linear phase_hcz (), its coefficient is a₁(0), a₂(0) ..., a_n(0), b₀(0), b₁(0) ... b_n(0);

(4) by G (z)=G_bs(z) G_hcz () asks for total transfer function G (z) of filter, and gather filtered signal and carry out power spectrumanalysis, gathers 1s filtered signal carry out power spectrumanalysis, if noise attentuation is lower than A in G (z) stopband with 2KHz_decay, increase K according to the relations of 2 times_decay, i.e. K_decay=2K_decay, re-start step (2)-step (4), try to achieve the numerator coefficients b of G (z)₀(n), b₁(n), b₂(n) ... b_n(n) and denominator coefficients a₁(n), a₂(n) ..., a_nn (), obtains the transfer function of filter