Index Home About Blog
From: rja14@cl.cam.ac.uk (Ross Anderson)
Date: 17 Jun 1994 13:43:28 GMT
Newsgroups: sci.crypt,alt.security,uk.telecom
Subject: A5 (Was: HACKING DIGITAL PHONES)

The GSM encryption algorithm, A5, is not much good. Its effective key
length is at most five bytes; and anyone with the time and energy to
look for faster attacks can find source code for it at the bottom of
this post.

The politics of all this is bizarre. Readers may recall that there was a
fuss last year about whether GSM phones could be exported to the Middle
East; the official line then was that A5 was too good for the likes of
Saddam Hussein.

However, a couple of weeks ago, they switched from saying that A5 was
too strong to disclose, to saying that it was too weak to disclose! The
government line now pleads that discussing it might harm export sales. 

Maybe all the fuss was just a ploy to get Saddam to buy A5 chips on the
black market; but Occam's razor suggests that we are really seeing the
results of the usual blundering, infighting and incompetence of bloated
government departments. 

Indeed, my spies inform me that there was a terrific row between the
NATO signals agencies in the mid 1980's over whether GSM encryption
should be strong or not. The Germans said it should be, as they shared
a long border with the Evil Empire; but the other countries didn't feel
this way, and the algorithm as now fielded is a French design.

A5 is a stream cipher, and the keystream is the xor of three clock
controlled registers. The clock control of each register is that
register's own middle bit, xor'ed with a threshold function of the
middle bits of all three registers (ie if two or more of the middle
bits are 1, then invert each of these bits; otherwise just use them as
they are). The register lengths are 19, 22 and 23, and all the feedback
polynomials are sparse.

Readers will note that there is a trivial 2^40 attack (guess the
contents of registers 1 and 2, work out register 3 from the keystream,
and then step on to check whether the guess was right). 2^40 trial
encryptions could take weeks on a workstation, but the low gate count
of the algorithm means that a Xilinx chip can easily be programmed to
do keysearch, and an A5 cracker might have a few dozen of these running
at maybe 2 keys per microsecond each. Of course, if all you want to do
is break the Royal Family's keys for sale to News International, then
software would do fine.

It is thus clear that A5 should be free of all export controls, just
like CDMF and the 40-bit versions of RC2 and RC4.

Indeed, there seems to be an even faster attack. As the clock control is
stop-go rather than 1-2, one would expect some kind of correlation
attack to be possible, and on June 3rd, Dr Simon Shepherd of Bradford
University was due to present an attack on A5 to an IEE colloquium in
London. However, his talk was spiked at the last minute by GCHQ, and
all we know about his attack is:

(a) that sparse matrix techniques are used to reconstruct the initial
    state (this was published as a `trailer' in the April 93 `Mobile
    Europe');

(b) that he used some of the tricks from my paper `Solving a class of
    stream ciphers' (Cryptologia XIV no 3 [July 90] pp 285 - 288) and
    from the follow-up paper `Divide and conquer attacks on certain
    classes of stream ciphers' by Ed Dawson and Andy Clark (Cryptologia
    XVIII no 1 [Jan 94] pp 25 - 40) (he mentioned this to me on the
    phone).

I believe that we have to stand up for academic freedom, and I hope that
placing A5 in the public domain will lead to the embargo on Simon's
paper being lifted.


Ross Anderson


APPENDIX - AN IMPLEMENTATION OF A5

The documentation we have, which arrived anonymously in two brown
envelopes, is incomplete; we do not know the feedback taps of registers
2 and 3, but we do know from the chip's gate count that they have at
most 6 feedback taps between them.

The following implementation of A5 is due to Mike Roe
<mrr@cl.cam.ac.uk>, and all comments and queries should be sent to him.



/*
 * In writing this program, I've had to guess a few pices of information:
 *
 * 1. Which bits of the key are loaded into which bits of the shift register
 * 2. Which order the frame sequence number is shifted into the SR (MSB
 *    first or LSB first)
 * 3. The position of the feedback taps on R2 and R3 (R1 is known).
 * 4. The position of the clock control taps. These are on the `middle' one, 
 *    I've assumed to be 9 on R1, 11 on R2, 11 on R3.
 */

/*
 * Look at the `middle' stage of each of the 3 shift registers.
 * Either 0, 1, 2 or 3 of these 3 taps will be set high.
 * If 0 or 1 or one of them are high, return true. This will cause each of
 * the middle taps to be inverted before being used as a clock control. In
 * all cases either 2 or 3 of the clock enable lines will be active. Thus,
 * at least two shift registers change on every clock-tick and the system
 * never becomes stuck.
 */

static int threshold(r1, r2, r3)
unsigned int r1;
unsigned int r2;
unsigned int r3;
{
int total;

  total = (((r1 >>  9) & 0x1) == 1) +
          (((r2 >> 11) & 0x1) == 1) +
          (((r3 >> 11) & 0x1) == 1);

  if (total > 1)
    return (0);
  else
    return (1);
}

unsigned long clock_r1(ctl, r1)
int ctl;
unsigned long r1;
{
unsigned long feedback;

 /*
  * Primitive polynomial x**19 + x**5 + x**2 + x + 1
  */

  ctl ^= ((r1 >> 9) & 0x1);
  if (ctl)
  {
    feedback = (r1 >> 18) ^ (r1 >> 17) ^ (r1 >> 16) ^ (r1 >> 13);
    r1 = (r1 << 1) & 0x7ffff;
    if (feedback & 0x01)
      r1 ^= 0x01;
  }
  return (r1);
}

unsigned long clock_r2(ctl, r2)
int ctl;
unsigned long r2;
{
unsigned long feedback;

  
 /*
  * Primitive polynomial x**22 + x**9 + x**5 + x + 1
  */   

  ctl ^= ((r2 >> 11) & 0x1);
  if (ctl)
  {
    feedback = (r2 >> 21) ^ (r2 >> 20) ^ (r2 >> 16) ^ (r2 >> 12);
    r2 = (r2 << 1) & 0x3fffff;
    if (feedback & 0x01)
      r2 ^= 0x01;
  }
  return (r2);
}

unsigned long clock_r3(ctl, r3)
int ctl;
unsigned long r3;
{
unsigned long feedback;

 /*
  * Primitive polynomial x**23 + x**5 + x**4 + x + 1
  */

  ctl ^= ((r3 >> 11) & 0x1);
  if (ctl)
  {
    feedback = (r3 >> 22) ^ (r3 >> 21) ^ (r3 >> 18) ^ (r3 >> 17);
    r3 = (r3 << 1) & 0x7fffff;
    if (feedback & 0x01)
      r3 ^= 0x01;
  }
  return (r3);
}

int keystream(key, frame, alice, bob)
unsigned char *key;   /* 64 bit session key              */
unsigned long frame;  /* 22 bit frame sequence number    */
unsigned char *alice; /* 114 bit Alice to Bob key stream */
unsigned char *bob;   /* 114 bit Bob to Alice key stream */
{
unsigned long r1;   /* 19 bit shift register */
unsigned long r2;   /* 22 bit shift register */
unsigned long r3;   /* 23 bit shift register */
int i;              /* counter for loops     */
int clock_ctl;      /* xored with clock enable on each shift register */
unsigned char *ptr; /* current position in keystream */
unsigned char byte; /* byte of keystream being assembled */
unsigned int bits;  /* number of bits of keystream in byte */
unsigned int bit;   /* bit output from keystream generator */

  /* Initialise shift registers from session key */

  r1 = (key[0] | (key[1] << 8) | (key[2] << 16) ) & 0x7ffff;
  r2 = ((key[2] >> 3) | (key[3] << 5) | (key[4] << 13) | (key[5] << 21)) &
0x3fffff;
  r3 = ((key[5] >> 1) | (key[6] << 7) | (key[7] << 15) ) & 0x7fffff;


  /* Merge frame sequence number into shift register state, by xor'ing it
   * into the feedback path
   */

  for (i=0;i<22;i++)
  {
    clock_ctl = threshold(r1, r2, r2);
    r1 = clock_r1(clock_ctl, r1);
    r2 = clock_r2(clock_ctl, r2);
    r3 = clock_r3(clock_ctl, r3);
    if (frame & 1)
    {
      r1 ^= 1;
      r2 ^= 1;
      r3 ^= 1;
    }
    frame = frame >> 1;
  }

  /* Run shift registers for 100 clock ticks to allow frame number to
   * be diffused into all the bits of the shift registers
   */

  for (i=0;i<100;i++)
  {
    clock_ctl = threshold(r1, r2, r2);
    r1 = clock_r1(clock_ctl, r1);
    r2 = clock_r2(clock_ctl, r2);
    r3 = clock_r3(clock_ctl, r3);
  }

  /* Produce 114 bits of Alice->Bob key stream */

  ptr = alice;
  bits = 0;
  byte = 0;
  for (i=0;i<114;i++)
  {
    clock_ctl = threshold(r1, r2, r2);
    r1 = clock_r1(clock_ctl, r1);
    r2 = clock_r2(clock_ctl, r2);
    r3 = clock_r3(clock_ctl, r3);

    bit = ((r1 >> 18) ^ (r2 >> 21) ^ (r3 >> 22)) & 0x01;
    byte = (byte << 1) | bit;
    bits++;
    if (bits == 8)
    {
      *ptr = byte;
      ptr++;
      bits = 0;
      byte = 0;
    }
  }
  if (bits)
    *ptr = byte;

  /* Run shift registers for another 100 bits to hide relationship between
   * Alice->Bob key stream and Bob->Alice key stream.
   */

  for (i=0;i<100;i++)
  {
    clock_ctl = threshold(r1, r2, r2);
    r1 = clock_r1(clock_ctl, r1);
    r2 = clock_r2(clock_ctl, r2);
    r3 = clock_r3(clock_ctl, r3);
  }

  /* Produce 114 bits of Bob->Alice key stream */

  ptr = bob;
  bits = 0;
  byte = 0;
  for (i=0;i<114;i++)
  {
    clock_ctl = threshold(r1, r2, r2);
    r1 = clock_r1(clock_ctl, r1);
    r2 = clock_r2(clock_ctl, r2);
    r3 = clock_r3(clock_ctl, r3);

    bit = ((r1 >> 18) ^ (r2 >> 21) ^ (r3 >> 22)) & 0x01;
    byte = (byte << 1) | bit;
    bits++;
    if (bits == 8)
    {
      *ptr = byte;
      ptr++;
      bits = 0;
      byte = 0;
    }
  }
  if (bits)
    *ptr = byte;
 
  return (0);

}

Index Home About Blog