Hi!
I'm trying to build a moving average filter and an error is occuring.
Basicaly, I'm using an accumulator to add the data present in
maf_reg(64) and subtract the data present in maf_reg(0). The addition is
perfectly working, but the subtraction isn't. The operation is between
lines 94 and 98.
CODE:
library ieee;
use ieee.std_logic_1164.all;
use ieee.std_logic_arith.all;
use ieee.std_logic_unsigned.all;
entity MAF6bits is
port(
clk : in std_logic;
rst : in std_logic;
sigIN : in std_logic;
-- bit stream
sigOUT : out std_logic_vector (5 downto 0)
-- average
);
end entity;
architecture behaviour of MAF6bits is
signal maf_reg : std_logic_vector (64 downto 0); -- registrador
signal acc : std_logic_vector (5 downto 0);
signal temp : std_logic_vector (1 downto 0);
begin
process( rst , clk )
begin
if ( rst='1' ) then
maf_reg <=
"00000000000000000000000000000000000000000000000000000000000000000";
acc <= "000000";
elsif ( clk'event and clk='1' ) then
maf_reg(64) <= sigIN;
maf_reg(63) <= maf_reg(64);
maf_reg(62) <= maf_reg(63);
maf_reg(61) <= maf_reg(62);
maf_reg(60) <= maf_reg(61);
maf_reg(59) <= maf_reg(60);
maf_reg(58) <= maf_reg(59);
maf_reg(57) <= maf_reg(58);
maf_reg(56) <= maf_reg(57);
maf_reg(55) <= maf_reg(56);
maf_reg(54) <= maf_reg(55);
maf_reg(53) <= maf_reg(54);
maf_reg(52) <= maf_reg(53);
maf_reg(51) <= maf_reg(52);
maf_reg(50) <= maf_reg(51);
maf_reg(49) <= maf_reg(50);
maf_reg(48) <= maf_reg(49);
maf_reg(47) <= maf_reg(48);
maf_reg(46) <= maf_reg(47);
maf_reg(45) <= maf_reg(46);
maf_reg(44) <= maf_reg(45);
maf_reg(43) <= maf_reg(44);
maf_reg(42) <= maf_reg(43);
maf_reg(41) <= maf_reg(42);
maf_reg(40) <= maf_reg(41);
maf_reg(39) <= maf_reg(40);
maf_reg(38) <= maf_reg(39);
maf_reg(37) <= maf_reg(38);
maf_reg(36) <= maf_reg(37);
maf_reg(35) <= maf_reg(36);
maf_reg(34) <= maf_reg(35);
maf_reg(33) <= maf_reg(34);
maf_reg(32) <= maf_reg(33);
maf_reg(31) <= maf_reg(32);
maf_reg(30) <= maf_reg(31);
maf_reg(29) <= maf_reg(30);
maf_reg(28) <= maf_reg(29);
maf_reg(27) <= maf_reg(28);
maf_reg(26) <= maf_reg(27);
maf_reg(25) <= maf_reg(26);
maf_reg(24) <= maf_reg(25);
maf_reg(23) <= maf_reg(24);
maf_reg(22) <= maf_reg(23);
maf_reg(21) <= maf_reg(22);
maf_reg(20) <= maf_reg(21);
maf_reg(19) <= maf_reg(20);
maf_reg(18) <= maf_reg(19);
maf_reg(17) <= maf_reg(18);
maf_reg(16) <= maf_reg(17);
maf_reg(15) <= maf_reg(16);
maf_reg(14) <= maf_reg(15);
maf_reg(13) <= maf_reg(14);
maf_reg(12) <= maf_reg(13);
maf_reg(11) <= maf_reg(12);
maf_reg(10) <= maf_reg(11);
maf_reg(9) <= maf_reg(10);
maf_reg(8) <= maf_reg(9);
maf_reg(7) <= maf_reg(8);
maf_reg(6) <= maf_reg(7);
maf_reg(5) <= maf_reg(6);
maf_reg(4) <= maf_reg(5);
maf_reg(3) <= maf_reg(4);
maf_reg(2) <= maf_reg(3);
maf_reg(1) <= maf_reg(2);
maf_reg(0) <= maf_reg(1);
if ( temp = "01" ) then
acc <= acc - "000001";
elsif ( temp = "10" ) then
acc <= acc + "000001";
end if;
end if;
temp <= maf_reg(64)&maf_reg(0);
end process;
sigOUT(5) <= acc(5);
sigOUT(4) <= acc(4);
sigOUT(3) <= acc(3);
sigOUT(2) <= acc(2);
sigOUT(1) <= acc(1);
sigOUT(0) <= acc(0);
end architecture;
1 | maf_reg(63) <= maf_reg(64); |
2 | maf_reg(62) <= maf_reg(63); |
3 | maf_reg(61) <= maf_reg(62); |
4 | maf_reg(60) <= maf_reg(61); |
5 | maf_reg(59) <= maf_reg(60); |
6 | :
|
7 | :
|
8 | maf_reg(2) <= maf_reg(3); |
9 | maf_reg(1) <= maf_reg(2); |
10 | maf_reg(0) <= maf_reg(1); |
You could cut down this description by a factor of 20 like this:
1 | for i in 0 to 63 loop |
2 | maf_reg(i) <= maf_reg(i+1); |
3 | end loop; |
You also could write:
1 | for i in 63 downto 0 loop |
2 | maf_reg(i) <= maf_reg(i+1); |
3 | end loop; |
Or take this line:
1 | maf_reg <= "00000000000000000000000000000000000000000000000000000000000000000"; |
With this slightly shorter and easier readable phrase:
1 | maf_reg <= (others=>'0'); |
William Marques wrote: > The addition is perfectly working, but the subtraction isn't. What does not work? What do you expect? And what happens instead? Maybe you have a little problem here: > use ieee.std_logic_unsigned.all;
:
Edited by Moderator
I think you could even so write the shift like:
1 | maf_reg(63 downto 0) <= maf_reg(64 downto 1); |
Depends on what is more obvious to you.
In the if statement:
if ( temp = "01" ) then
acc <= acc - "000001";
elsif ( temp = "10" ) then
acc <= acc + "000001";
end if;
I intend to add '1' to the accumulator acc if there is '1' in
maf_reg(64) and '0' in maf_reg(0) and subtract '1' if there is a '0' in
maf_reg(64) and '1' in maf_reg(0). The problem is that the summation is
working and the accumulator increases its value by 1 but when the second
condition occurs the accumulator doesn't subtract 1. Instead, it remains
with its value. So it's just increasing (when temp = "01") and never
decreasing (when temp = "10").
Firstly, thanks.
In the if statement:
if ( temp = "01" ) then
acc <= acc - "000001";
elsif ( temp = "10" ) then
acc <= acc + "000001";
end if;
I intend to add '1' to the accumulator acc if there is '1' in
maf_reg(64) and '0' in maf_reg(0) and subtract '1' if there is a '0' in
maf_reg(64) and '1' in maf_reg(0). The problem is that the summation is
working and the accumulator increases its value by 1 but when the second
condition occurs the accumulator doesn't subtract 1. Instead, it remains
with its value.
So it's just increasing (when temp = "10") and never decreasing (when
temp = "01").
Watch this topic
Disable multi-page viewPlease log in before posting. Registration is free and takes only a minute.
Existing account
Do you have a Google/GoogleMail account? No registration required!
Log in with Google account
Log in with Google account
No account? Register here.